prepare_audio_1_tfdata_ver2.ipynb

{
  "cells": [
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "6ypmOr_03EHP"
      },
      "source": [
        "Here's what this notebook does, step-by-step:\n",
        "\n",
        "1. Install the `montreal-force-aligner` (MFA) library using conda.\n",
        "2. Download the Viet Bible Vox dataset from Huggingface.\n",
        "3. Generate a normalized transcript file for each .mp3 file in the dataset.\n",
        "4. Train an MFA model, then align speech and phonemes (creating a timestamp for each phoneme).\n",
        "5. Generate a TensorFlow dataset (tfrecord files) that is ready for training TTS models."
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 1,
      "metadata": {
        "id": "SwhMuEANS0v4"
      },
      "outputs": [],
      "source": [
        "# %pip install -Uqq tensorflow numpy torch tqdm librosa numba regex"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 2,
      "metadata": {},
      "outputs": [
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "/home/billgates/lightspeed/data\n"
          ]
        }
      ],
      "source": [
        "# ./data is our working directory\n",
        "# !mkdir -p data\n",
        "%cd data"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 3,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "euumE7pMw6j8",
        "outputId": "0c54a090-f8c5-4729-fb86-f3e47bf6a3c8"
      },
      "outputs": [],
      "source": [
        "#### INSTALL MFA  ####\n",
        "# !wget https://repo.anaconda.com/miniconda/Miniconda3-py311_23.5.2-0-Linux-x86_64.sh -qO ./miniconda.sh\n",
        "# # !wget https://repo.anaconda.com/miniconda/Miniconda3-latest-MacOSX-arm64.sh -qO $PWD/miniconda.sh # for Apple M1\n",
        "# !bash ./miniconda.sh -b -p ./miniconda\n",
        "# !source ./miniconda/bin/activate && conda create -n aligner -c conda-forge montreal-forced-aligner=2.2.15 -y --quiet\n",
        "# if done install\n",
        "# !source ./miniconda/bin/activate aligner"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 4,
      "metadata": {},
      "outputs": [
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "/home/billgates/lightspeed/data\n"
          ]
        }
      ],
      "source": [
        "!pwd"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 6,
      "metadata": {
        "id": "1whkRTu5NBam"
      },
      "outputs": [
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "2120\n",
            "2120\n"
          ]
        }
      ],
      "source": [
        "# !git lfs install # LFS is needed to download the data\n",
        "# !git clone https://huggingface.co/datasets/ntt123/VietBibleVox\n",
        "# !cd audio_1; unzip -qq audio_1.zip\n",
        "!cd audio_1_experiment/audio_1_ver2; ls *.wav | wc -l\n",
        "!cd audio_1_experiment/audio_1_ver2; ls *.txt | wc -l"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 17,
      "metadata": {
        "id": "sPFR2tbJV1au"
      },
      "outputs": [],
      "source": [
        "import regex\n",
        "import unicodedata\n",
        "from pathlib import Path"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 18,
      "metadata": {
        "id": "QgaPWypIWDQF"
      },
      "outputs": [],
      "source": [
        "vietnamese_characters = [\n",
        "    'a', 'à', 'á', 'ả', 'ã', 'ạ',\n",
        "    'ă', 'ằ', 'ắ', 'ẳ', 'ẵ', 'ặ',\n",
        "    'â', 'ầ', 'ấ', 'ẩ', 'ẫ', 'ậ',\n",
        "    'e', 'è', 'é', 'ẻ', 'ẽ', 'ẹ',\n",
        "    'ê', 'ề', 'ế', 'ể', 'ễ', 'ệ',\n",
        "    'i', 'ì', 'í', 'ỉ', 'ĩ', 'ị',\n",
        "    'o', 'ò', 'ó', 'ỏ', 'õ', 'ọ',\n",
        "    'ô', 'ồ', 'ố', 'ổ', 'ỗ', 'ộ',\n",
        "    'ơ', 'ờ', 'ớ', 'ở', 'ỡ', 'ợ',\n",
        "    'u', 'ù', 'ú', 'ủ', 'ũ', 'ụ',\n",
        "    'ư', 'ừ', 'ứ', 'ử', 'ữ', 'ự',\n",
        "    'y', 'ỳ', 'ý', 'ỷ', 'ỹ', 'ỵ',\n",
        "    'b', 'c', 'd', 'đ', 'g', 'h', \n",
        "    'k', 'l', 'm', 'n', 'p', 'q', \n",
        "    'r', 's', 't', 'v', 'x'\n",
        "]\n",
        "alphabet = \"\".join(vietnamese_characters)\n",
        "space_re = regex.compile(r\"\\s+\")\n",
        "number_re = regex.compile(\"([0-9]+)\")\n",
        "digits = [\"không\", \"một\", \"hai\", \"ba\", \"bốn\", \"năm\", \"sáu\", \"bảy\", \"tám\", \"chín\"]\n",
        "num_re = regex.compile(r\"([0-9.,]*[0-9])\")\n",
        "keep_text_and_num_re = regex.compile(rf'[^\\s{alphabet}.,0-9]')\n",
        "keep_text_re = regex.compile(rf'[^\\s{alphabet}]')"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 19,
      "metadata": {
        "id": "jXeBft4Ctk9p"
      },
      "outputs": [],
      "source": [
        "def read_number(num: str) -> str:\n",
        "    if len(num) == 1:\n",
        "        return digits[int(num)]\n",
        "    elif len(num) == 2 and num.isdigit():\n",
        "        n = int(num)\n",
        "        end = digits[n % 10]\n",
        "        if n == 10:\n",
        "            return \"mười\"\n",
        "        if n % 10 == 5:\n",
        "            end = \"lăm\"\n",
        "        if n % 10 == 0:\n",
        "            return digits[n // 10] + \" mươi\"\n",
        "        elif n < 20:\n",
        "            return \"mười \" + end\n",
        "        else:\n",
        "            if n % 10 == 1:\n",
        "                end = \"mốt\"\n",
        "            return digits[n // 10] + \" mươi \" + end\n",
        "    elif len(num) == 3 and num.isdigit():\n",
        "        n = int(num)\n",
        "        if n % 100 == 0:\n",
        "            return digits[n // 100] + \" trăm\"\n",
        "        elif num[1] == \"0\":\n",
        "            return digits[n // 100] + \" trăm lẻ \" + digits[n % 100]\n",
        "        else:\n",
        "            return digits[n // 100] + \" trăm \" + read_number(num[1:])\n",
        "    elif len(num) >= 4 and len(num) <= 6 and num.isdigit():\n",
        "        n = int(num)\n",
        "        n1 = n // 1000\n",
        "        return read_number(str(n1)) + \" ngàn \" + read_number(num[-3:])\n",
        "    elif \",\" in num:\n",
        "        n1, n2 = num.split(\",\")\n",
        "        return read_number(n1) + \" phẩy \" + read_number(n2)\n",
        "    elif \".\" in num:\n",
        "        parts = num.split(\".\")\n",
        "        if len(parts) == 2:\n",
        "            if parts[1] == \"000\":\n",
        "                return read_number(parts[0]) + \" ngàn\"\n",
        "            elif parts[1].startswith(\"00\"):\n",
        "                end = digits[int(parts[1][2:])]\n",
        "                return read_number(parts[0]) + \" ngàn lẻ \" + end\n",
        "            else:\n",
        "                return read_number(parts[0]) + \" ngàn \" + read_number(parts[1])\n",
        "        elif len(parts) == 3:\n",
        "            return (\n",
        "                read_number(parts[0])\n",
        "                + \" triệu \"\n",
        "                + read_number(parts[1])\n",
        "                + \" ngàn \"\n",
        "                + read_number(parts[2])\n",
        "            )\n",
        "    return num"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 20,
      "metadata": {
        "id": "5taIbrZ5WGX8"
      },
      "outputs": [],
      "source": [
        "def normalize_text(x):\n",
        "    x = unicodedata.normalize('NFKC', x)\n",
        "    x = x.lower()\n",
        "    x = num_re.sub(r\" \\1 \", x)\n",
        "    x = keep_text_and_num_re.sub(\" \", x)\n",
        "    words = x.split()\n",
        "    words = [ read_number(w) if num_re.fullmatch(w) else w for w in words ]\n",
        "    x = \" \".join(words)\n",
        "    x = keep_text_re.sub(\" \", x)\n",
        "    x = space_re.sub(\" \", x)\n",
        "    x = x.strip()\n",
        "    return x"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 21,
      "metadata": {
        "id": "-QWNUkoVV6WF"
      },
      "outputs": [],
      "source": [
        "all_text = []\n",
        "for fp in sorted(Path(\"audio_1_experiment/audio_1_ver2\").glob(\"*.txt\")):\n",
        "    with open(fp, \"r\", encoding=\"utf-8\") as f:\n",
        "        text = f.read()\n",
        "        # print(text)\n",
        "        text = normalize_text(text)\n",
        "        all_text.append(text)\n",
        "    # override the text file\n",
        "    with open(fp, \"w\", encoding=\"utf-8\") as f:\n",
        "        f.write(text)\n",
        "all_words = sorted(set((\" \".join(all_text)).split()))"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 22,
      "metadata": {
        "id": "JFZDqnMzcH5g"
      },
      "outputs": [],
      "source": [
        "with open(\"lexicon.txt\", \"w\") as f:\n",
        "    for w in all_words:\n",
        "        w = w.strip()\n",
        "        p = list(w)\n",
        "        p = \" \".join(p)\n",
        "        f.write(f\"{w}\\t{p}\\n\")"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 23,
      "metadata": {},
      "outputs": [
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "[Errno 2] No such file or directory: 'data'\n",
            "/home/billgates/lightspeed/data\n"
          ]
        }
      ],
      "source": [
        "%cd data"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 24,
      "metadata": {},
      "outputs": [
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "Please be aware that you are running an alpha version of MFA. If you would like to install a more stable version, please visit https://montreal-forced-aligner.readthedocs.io/en/latest/installation.html#installing-older-versions-of-mfa\n",
            "\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m Setting up corpus information\u001b[33m...\u001b[0m                                      \n",
            "\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m Loading corpus from source files\u001b[33m...\u001b[0m                                   \n",
            "\u001b[2K\u001b[35m 100%\u001b[0m \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m1,711/100 \u001b[0m [ \u001b[33m0:00:00\u001b[0m < \u001b[36m0:00:00\u001b[0m , \u001b[31m9,116 it/s\u001b[0m ]\n",
            "\u001b[?25h\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m Found \u001b[1;36m1\u001b[0m speaker across \u001b[1;36m2120\u001b[0m files, average number of utterances per   \n",
            "\u001b[2;36m \u001b[0m         speaker: \u001b[1;36m2120.0\u001b[0m                                                       \n",
            "\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m Initializing multiprocessing jobs\u001b[33m...\u001b[0m                                  \n",
            "\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m Normalizing text\u001b[33m...\u001b[0m                                                   \n",
            "\u001b[2K\u001b[35m  38%\u001b[0m \u001b[91m━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m811/2,120 \u001b[0m [ \u001b[33m0:00:02\u001b[0m < \u001b[36m-:--:--\u001b[0m , \u001b[31m? it/s\u001b[0m ]\n",
            "\u001b[?25h\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m Generating MFCCs\u001b[33m...\u001b[0m                                                   \n",
            "\u001b[2K\u001b[35m  96%\u001b[0m \u001b[91m━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[91m╸\u001b[0m\u001b[90m━\u001b[0m \u001b[32m2,040/2,120 \u001b[0m [ \u001b[33m0:00:04\u001b[0m < \u001b[36m0:00:01\u001b[0m , \u001b[31m938 it/s\u001b[0m ]\n",
            "\u001b[?25h\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m Calculating CMVN\u001b[33m...\u001b[0m                                                   \n",
            "\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m Generating final features\u001b[33m...\u001b[0m                                          \n",
            "\u001b[2K\u001b[35m  83%\u001b[0m \u001b[91m━━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[91m╸\u001b[0m\u001b[90m━━━━\u001b[0m \u001b[32m1,755/2,120 \u001b[0m [ \u001b[33m0:00:02\u001b[0m < \u001b[36m0:00:01\u001b[0m , \u001b[31m4,563 it/s\u001b[0m ]\n",
            "\u001b[?25h\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m Creating corpus split\u001b[33m...\u001b[0m                                              \n",
            "\u001b[2K\u001b[35m   0%\u001b[0m \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m0/2,120 \u001b[0m [ \u001b[33m0:00:02\u001b[0m < \u001b[36m-:--:--\u001b[0m , \u001b[31m? it/s\u001b[0m ]\n",
            "\u001b[?25h\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m Filtering utterances for training\u001b[33m...\u001b[0m                                  \n",
            "\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m Initializing training for monophone\u001b[33m...\u001b[0m                                \n",
            "\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m Compiling training graphs\u001b[33m...\u001b[0m                                          \n",
            "\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m Generating initial alignments\u001b[33m...\u001b[0m                                      \n",
            "\u001b[2K\u001b[35m 100%\u001b[0m \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m2,128/2,120 \u001b[0m [ \u001b[33m0:00:00\u001b[0m < \u001b[36m0:00:00\u001b[0m , \u001b[31m3,723 it/s\u001b[0m ]1m3,719 it/s\u001b[0m ]\n",
            "\u001b[?25h\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m Initialization complete!                                              \n",
            "\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m monophone - Iteration \u001b[1;36m1\u001b[0m of \u001b[1;36m40\u001b[0m                                         \n",
            "\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m Generating alignments\u001b[33m...\u001b[0m                                              \n",
            "\u001b[2K\u001b[35m   1%\u001b[0m \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m12/2,120 \u001b[0m [ \u001b[33m0:00:09\u001b[0m < \u001b[36m0:15:41\u001b[0m , \u001b[31m2 it/s\u001b[0m ]\n",
            "\u001b[?25h\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m Accumulating statistics\u001b[33m...\u001b[0m                                            \n",
            "\u001b[2K\u001b[35m   0%\u001b[0m \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m0/2,120 \u001b[0m [ \u001b[33m0:00:02\u001b[0m < \u001b[36m-:--:--\u001b[0m , \u001b[31m? it/s\u001b[0m ]\n",
            "\u001b[?25h\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m monophone - Iteration \u001b[1;36m2\u001b[0m of \u001b[1;36m40\u001b[0m                                         \n",
            "\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m Generating alignments\u001b[33m...\u001b[0m                                              \n",
            "\u001b[2K\u001b[35m  17%\u001b[0m \u001b[91m━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m365/2,120 \u001b[0m [ \u001b[33m0:00:11\u001b[0m < \u001b[36m0:00:44\u001b[0m , \u001b[31m41 it/s\u001b[0m ]\n",
            "\u001b[?25h\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m Accumulating statistics\u001b[33m...\u001b[0m                                            \n",
            "\u001b[2K\u001b[35m   0%\u001b[0m \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m0/2,120 \u001b[0m [ \u001b[33m0:00:02\u001b[0m < \u001b[36m-:--:--\u001b[0m , \u001b[31m? it/s\u001b[0m ]\n",
            "\u001b[?25h\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m monophone - Iteration \u001b[1;36m3\u001b[0m of \u001b[1;36m40\u001b[0m                                         \n",
            "\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m Generating alignments\u001b[33m...\u001b[0m                                              \n",
            "\u001b[2K\u001b[35m  99%\u001b[0m \u001b[91m━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[91m╸\u001b[0m \u001b[32m2,104/2,120 \u001b[0m [ \u001b[33m0:00:07\u001b[0m < \u001b[36m0:00:01\u001b[0m , \u001b[31m436 it/s\u001b[0m ]t/s\u001b[0m ]\n",
            "\u001b[?25h\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m Accumulating statistics\u001b[33m...\u001b[0m                                            \n",
            "\u001b[2K\u001b[35m 100%\u001b[0m \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m2,126/2,120 \u001b[0m [ \u001b[33m0:00:00\u001b[0m < \u001b[36m0:00:00\u001b[0m , \u001b[31m7,696 it/s\u001b[0m ]1m7,372 it/s\u001b[0m ]\n",
            "\u001b[?25h\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m monophone - Iteration \u001b[1;36m4\u001b[0m of \u001b[1;36m40\u001b[0m                                         \n",
            "\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m Generating alignments\u001b[33m...\u001b[0m                                              \n",
            "\u001b[2K\u001b[35m  99%\u001b[0m \u001b[91m━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[91m╸\u001b[0m \u001b[32m2,103/2,120 \u001b[0m [ \u001b[33m0:00:04\u001b[0m < \u001b[36m0:00:01\u001b[0m , \u001b[31m933 it/s\u001b[0m ]t/s\u001b[0m ]\n",
            "\u001b[?25h\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m Accumulating statistics\u001b[33m...\u001b[0m                                            \n",
            "\u001b[2K\u001b[35m  71%\u001b[0m \u001b[91m━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━\u001b[0m \u001b[32m1,508/2,120 \u001b[0m [ \u001b[33m0:00:02\u001b[0m < \u001b[36m0:00:01\u001b[0m , \u001b[31m8,686 it/s\u001b[0m ]\n",
            "\u001b[?25h\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m monophone - Iteration \u001b[1;36m5\u001b[0m of \u001b[1;36m40\u001b[0m                                         \n",
            "\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m Generating alignments\u001b[33m...\u001b[0m                                              \n",
            "\u001b[2K\u001b[35m  99%\u001b[0m \u001b[91m━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[91m╸\u001b[0m \u001b[32m2,099/2,120 \u001b[0m [ \u001b[33m0:00:04\u001b[0m < \u001b[36m0:00:01\u001b[0m , \u001b[31m1,065 it/s\u001b[0m ]t/s\u001b[0m ]\n",
            "\u001b[?25h\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m Accumulating statistics\u001b[33m...\u001b[0m                                            \n",
            "\u001b[2K\u001b[35m  69%\u001b[0m \u001b[91m━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━\u001b[0m \u001b[32m1,473/2,120 \u001b[0m [ \u001b[33m0:00:02\u001b[0m < \u001b[36m0:00:01\u001b[0m , \u001b[31m8,358 it/s\u001b[0m ]\n",
            "\u001b[?25h\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m monophone - Iteration \u001b[1;36m6\u001b[0m of \u001b[1;36m40\u001b[0m                                         \n",
            "\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m Generating alignments\u001b[33m...\u001b[0m                                              \n",
            "\u001b[2K\u001b[35m  98%\u001b[0m \u001b[91m━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[91m╸\u001b[0m \u001b[32m2,083/2,120 \u001b[0m [ \u001b[33m0:00:05\u001b[0m < \u001b[36m0:00:01\u001b[0m , \u001b[31m648 it/s\u001b[0m ]t/s\u001b[0m ]\n",
            "\u001b[?25h\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m Accumulating statistics\u001b[33m...\u001b[0m                                            \n",
            "\u001b[2K\u001b[35m  92%\u001b[0m \u001b[91m━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[91m╸\u001b[0m\u001b[90m━━\u001b[0m \u001b[32m1,949/2,120 \u001b[0m [ \u001b[33m0:00:02\u001b[0m < \u001b[36m0:00:01\u001b[0m , \u001b[31m7,007 it/s\u001b[0m ]\n",
            "\u001b[?25h\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m monophone - Iteration \u001b[1;36m7\u001b[0m of \u001b[1;36m40\u001b[0m                                         \n",
            "\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m Generating alignments\u001b[33m...\u001b[0m                                              \n",
            "\u001b[2K\u001b[35m  98%\u001b[0m \u001b[91m━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m \u001b[32m2,076/2,120 \u001b[0m [ \u001b[33m0:00:04\u001b[0m < \u001b[36m0:00:01\u001b[0m , \u001b[31m792 it/s\u001b[0m ]t/s\u001b[0m ]\n",
            "\u001b[?25h\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m Accumulating statistics\u001b[33m...\u001b[0m                                            \n",
            "\u001b[2K\u001b[35m  71%\u001b[0m \u001b[91m━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[91m╸\u001b[0m\u001b[90m━━━━━━━\u001b[0m \u001b[32m1,510/2,120 \u001b[0m [ \u001b[33m0:00:02\u001b[0m < \u001b[36m0:00:01\u001b[0m , \u001b[31m8,738 it/s\u001b[0m ]\n",
            "\u001b[?25h\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m monophone - Iteration \u001b[1;36m8\u001b[0m of \u001b[1;36m40\u001b[0m                                         \n",
            "\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m Generating alignments\u001b[33m...\u001b[0m                                              \n",
            "\u001b[2K\u001b[35m  98%\u001b[0m \u001b[91m━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[91m╸\u001b[0m \u001b[32m2,083/2,120 \u001b[0m [ \u001b[33m0:00:04\u001b[0m < \u001b[36m0:00:01\u001b[0m , \u001b[31m765 it/s\u001b[0m ]t/s\u001b[0m ]\n",
            "\u001b[?25h\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m Accumulating statistics\u001b[33m...\u001b[0m                                            \n",
            "\u001b[2K\u001b[35m  79%\u001b[0m \u001b[91m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[91m╸\u001b[0m\u001b[90m━━━━━\u001b[0m \u001b[32m1,674/2,120 \u001b[0m [ \u001b[33m0:00:02\u001b[0m < \u001b[36m0:00:01\u001b[0m , \u001b[31m8,284 it/s\u001b[0m ]\n",
            "\u001b[?25h\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m monophone - Iteration \u001b[1;36m9\u001b[0m of \u001b[1;36m40\u001b[0m                                         \n",
            "\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m Generating alignments\u001b[33m...\u001b[0m                                              \n",
            "\u001b[2K\u001b[35m  99%\u001b[0m \u001b[91m━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[91m╸\u001b[0m \u001b[32m2,103/2,120 \u001b[0m [ \u001b[33m0:00:05\u001b[0m < \u001b[36m0:00:01\u001b[0m , \u001b[31m699 it/s\u001b[0m ]t/s\u001b[0m ]\n",
            "\u001b[?25h\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m Accumulating statistics\u001b[33m...\u001b[0m                                            \n",
            "\u001b[2K\u001b[35m  68%\u001b[0m \u001b[91m━━━━━━━━━━━━━━━━━\u001b[0m\u001b[91m╸\u001b[0m\u001b[90m━━━━━━━━\u001b[0m \u001b[32m1,445/2,120 \u001b[0m [ \u001b[33m0:00:02\u001b[0m < \u001b[36m0:00:01\u001b[0m , \u001b[31m8,485 it/s\u001b[0m ]\n",
            "\u001b[?25h\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m monophone - Iteration \u001b[1;36m10\u001b[0m of \u001b[1;36m40\u001b[0m                                        \n",
            "\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m Generating alignments\u001b[33m...\u001b[0m                                              \n",
            "\u001b[2K\u001b[35m  97%\u001b[0m \u001b[91m━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m \u001b[32m2,064/2,120 \u001b[0m [ \u001b[33m0:00:04\u001b[0m < \u001b[36m0:00:01\u001b[0m , \u001b[31m759 it/s\u001b[0m ]t/s\u001b[0m ]\n",
            "\u001b[?25h\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m Accumulating statistics\u001b[33m...\u001b[0m                                            \n",
            "\u001b[2K\u001b[35m  69%\u001b[0m \u001b[91m━━━━━━━━━━━━━━━━━\u001b[0m\u001b[91m╸\u001b[0m\u001b[90m━━━━━━━━\u001b[0m \u001b[32m1,454/2,120 \u001b[0m [ \u001b[33m0:00:02\u001b[0m < \u001b[36m0:00:01\u001b[0m , \u001b[31m8,440 it/s\u001b[0m ]\n",
            "\u001b[?25h\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m monophone - Iteration \u001b[1;36m11\u001b[0m of \u001b[1;36m40\u001b[0m                                        \n",
            "\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m Accumulating statistics\u001b[33m...\u001b[0m                                            \n",
            "\u001b[2K\u001b[35m  69%\u001b[0m \u001b[91m━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━\u001b[0m \u001b[32m1,473/2,120 \u001b[0m [ \u001b[33m0:00:02\u001b[0m < \u001b[36m0:00:01\u001b[0m , \u001b[31m8,279 it/s\u001b[0m ]\n",
            "\u001b[?25h\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m monophone - Iteration \u001b[1;36m12\u001b[0m of \u001b[1;36m40\u001b[0m                                        \n",
            "\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m Generating alignments\u001b[33m...\u001b[0m                                              \n",
            "\u001b[2K\u001b[35m 100%\u001b[0m \u001b[91m━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[91m╸\u001b[0m \u001b[32m2,112/2,120 \u001b[0m [ \u001b[33m0:00:04\u001b[0m < \u001b[36m0:00:01\u001b[0m , \u001b[31m749 it/s\u001b[0m ]t/s\u001b[0m ]\n",
            "\u001b[?25h\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m Accumulating statistics\u001b[33m...\u001b[0m                                            \n",
            "\u001b[2K\u001b[35m  70%\u001b[0m \u001b[91m━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━\u001b[0m \u001b[32m1,475/2,120 \u001b[0m [ \u001b[33m0:00:02\u001b[0m < \u001b[36m0:00:01\u001b[0m , \u001b[31m8,517 it/s\u001b[0m ]\n",
            "\u001b[?25h\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m monophone - Iteration \u001b[1;36m13\u001b[0m of \u001b[1;36m40\u001b[0m                                        \n",
            "\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m Accumulating statistics\u001b[33m...\u001b[0m                                            \n",
            "\u001b[2K\u001b[35m  67%\u001b[0m \u001b[91m━━━━━━━━━━━━━━━━━\u001b[0m\u001b[91m╸\u001b[0m\u001b[90m━━━━━━━━\u001b[0m \u001b[32m1,429/2,120 \u001b[0m [ \u001b[33m0:00:02\u001b[0m < \u001b[36m0:00:01\u001b[0m , \u001b[31m8,255 it/s\u001b[0m ]\n",
            "\u001b[?25h\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m monophone - Iteration \u001b[1;36m14\u001b[0m of \u001b[1;36m40\u001b[0m                                        \n",
            "\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m Generating alignments\u001b[33m...\u001b[0m                                              \n",
            "\u001b[2K\u001b[35m  97%\u001b[0m \u001b[91m━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m \u001b[32m2,065/2,120 \u001b[0m [ \u001b[33m0:00:04\u001b[0m < \u001b[36m0:00:01\u001b[0m , \u001b[31m794 it/s\u001b[0m ]t/s\u001b[0m ]\n",
            "\u001b[?25h\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m Accumulating statistics\u001b[33m...\u001b[0m                                            \n",
            "\u001b[2K\u001b[35m  99%\u001b[0m \u001b[91m━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[91m╸\u001b[0m \u001b[32m2,093/2,120 \u001b[0m [ \u001b[33m0:00:02\u001b[0m < \u001b[36m0:00:01\u001b[0m , \u001b[31m7,057 it/s\u001b[0m ]t/s\u001b[0m ]\n",
            "\u001b[?25h\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m monophone - Iteration \u001b[1;36m15\u001b[0m of \u001b[1;36m40\u001b[0m                                        \n",
            "\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m Accumulating statistics\u001b[33m...\u001b[0m                                            \n",
            "\u001b[2K\u001b[35m  67%\u001b[0m \u001b[91m━━━━━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━\u001b[0m \u001b[32m1,424/2,120 \u001b[0m [ \u001b[33m0:00:02\u001b[0m < \u001b[36m0:00:01\u001b[0m , \u001b[31m8,147 it/s\u001b[0m ]\n",
            "\u001b[?25h\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m monophone - Iteration \u001b[1;36m16\u001b[0m of \u001b[1;36m40\u001b[0m                                        \n",
            "\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m Generating alignments\u001b[33m...\u001b[0m                                              \n",
            "\u001b[2K\u001b[35m  97%\u001b[0m \u001b[91m━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m \u001b[32m2,051/2,120 \u001b[0m [ \u001b[33m0:00:04\u001b[0m < \u001b[36m0:00:01\u001b[0m , \u001b[31m823 it/s\u001b[0m ]t/s\u001b[0m ]\n",
            "\u001b[?25h\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m Accumulating statistics\u001b[33m...\u001b[0m                                            \n",
            "\u001b[2K\u001b[35m  93%\u001b[0m \u001b[91m━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━\u001b[0m \u001b[32m1,962/2,120 \u001b[0m [ \u001b[33m0:00:02\u001b[0m < \u001b[36m0:00:01\u001b[0m , \u001b[31m6,619 it/s\u001b[0m ]\n",
            "\u001b[?25h\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m monophone - Iteration \u001b[1;36m17\u001b[0m of \u001b[1;36m40\u001b[0m                                        \n",
            "\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m Accumulating statistics\u001b[33m...\u001b[0m                                            \n",
            "\u001b[2K\u001b[35m  67%\u001b[0m \u001b[91m━━━━━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━\u001b[0m \u001b[32m1,411/2,120 \u001b[0m [ \u001b[33m0:00:02\u001b[0m < \u001b[36m0:00:01\u001b[0m , \u001b[31m8,210 it/s\u001b[0m ]\n",
            "\u001b[?25h\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m monophone - Iteration \u001b[1;36m18\u001b[0m of \u001b[1;36m40\u001b[0m                                        \n",
            "\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m Generating alignments\u001b[33m...\u001b[0m                                              \n",
            "\u001b[2K\u001b[35m 100%\u001b[0m \u001b[91m━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[91m╸\u001b[0m \u001b[32m2,116/2,120 \u001b[0m [ \u001b[33m0:00:04\u001b[0m < \u001b[36m0:00:01\u001b[0m , \u001b[31m800 it/s\u001b[0m ]t/s\u001b[0m ]\n",
            "\u001b[?25h\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m Accumulating statistics\u001b[33m...\u001b[0m                                            \n",
            "\u001b[2K\u001b[35m  68%\u001b[0m \u001b[91m━━━━━━━━━━━━━━━━━\u001b[0m\u001b[91m╸\u001b[0m\u001b[90m━━━━━━━━\u001b[0m \u001b[32m1,449/2,120 \u001b[0m [ \u001b[33m0:00:02\u001b[0m < \u001b[36m0:00:01\u001b[0m , \u001b[31m8,172 it/s\u001b[0m ]\n",
            "\u001b[?25h\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m monophone - Iteration \u001b[1;36m19\u001b[0m of \u001b[1;36m40\u001b[0m                                        \n",
            "\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m Accumulating statistics\u001b[33m...\u001b[0m                                            \n",
            "\u001b[2K\u001b[35m  67%\u001b[0m \u001b[91m━━━━━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━\u001b[0m \u001b[32m1,413/2,120 \u001b[0m [ \u001b[33m0:00:02\u001b[0m < \u001b[36m0:00:01\u001b[0m , \u001b[31m8,095 it/s\u001b[0m ]\n",
            "\u001b[?25h\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m monophone - Iteration \u001b[1;36m20\u001b[0m of \u001b[1;36m40\u001b[0m                                        \n",
            "\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m Generating alignments\u001b[33m...\u001b[0m                                              \n",
            "\u001b[2K\u001b[35m  96%\u001b[0m \u001b[91m━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[91m╸\u001b[0m\u001b[90m━\u001b[0m \u001b[32m2,044/2,120 \u001b[0m [ \u001b[33m0:00:04\u001b[0m < \u001b[36m0:00:01\u001b[0m , \u001b[31m822 it/s\u001b[0m ]\n",
            "\u001b[?25h\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m Accumulating statistics\u001b[33m...\u001b[0m                                            \n",
            "\u001b[2K\u001b[35m 100%\u001b[0m \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m2,132/2,120 \u001b[0m [ \u001b[33m0:00:00\u001b[0m < \u001b[36m0:00:00\u001b[0m , \u001b[31m7,577 it/s\u001b[0m ]1m7,991 it/s\u001b[0m ]\n",
            "\u001b[?25h\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m monophone - Iteration \u001b[1;36m21\u001b[0m of \u001b[1;36m40\u001b[0m                                        \n",
            "\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m Accumulating statistics\u001b[33m...\u001b[0m                                            \n",
            "\u001b[2K\u001b[35m 100%\u001b[0m \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m2,132/2,120 \u001b[0m [ \u001b[33m0:00:00\u001b[0m < \u001b[36m0:00:00\u001b[0m , \u001b[31m7,547 it/s\u001b[0m ]1m7,496 it/s\u001b[0m ]\n",
            "\u001b[?25h\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m monophone - Iteration \u001b[1;36m22\u001b[0m of \u001b[1;36m40\u001b[0m                                        \n",
            "\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m Accumulating statistics\u001b[33m...\u001b[0m                                            \n",
            "\u001b[2K\u001b[35m 100%\u001b[0m \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m2,130/2,120 \u001b[0m [ \u001b[33m0:00:00\u001b[0m < \u001b[36m0:00:00\u001b[0m , \u001b[31m7,952 it/s\u001b[0m ]1m8,207 it/s\u001b[0m ]\n",
            "\u001b[?25h\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m monophone - Iteration \u001b[1;36m23\u001b[0m of \u001b[1;36m40\u001b[0m                                        \n",
            "\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m Generating alignments\u001b[33m...\u001b[0m                                              \n",
            "\u001b[2K\u001b[35m  97%\u001b[0m \u001b[91m━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m \u001b[32m2,063/2,120 \u001b[0m [ \u001b[33m0:00:04\u001b[0m < \u001b[36m0:00:01\u001b[0m , \u001b[31m789 it/s\u001b[0m ]t/s\u001b[0m ]\n",
            "\u001b[?25h\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m Accumulating statistics\u001b[33m...\u001b[0m                                            \n",
            "\u001b[2K\u001b[35m  68%\u001b[0m \u001b[91m━━━━━━━━━━━━━━━━━\u001b[0m\u001b[91m╸\u001b[0m\u001b[90m━━━━━━━━\u001b[0m \u001b[32m1,437/2,120 \u001b[0m [ \u001b[33m0:00:02\u001b[0m < \u001b[36m0:00:01\u001b[0m , \u001b[31m8,279 it/s\u001b[0m ]\n",
            "\u001b[?25h\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m monophone - Iteration \u001b[1;36m24\u001b[0m of \u001b[1;36m40\u001b[0m                                        \n",
            "\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m Accumulating statistics\u001b[33m...\u001b[0m                                            \n",
            "\u001b[2K\u001b[35m 100%\u001b[0m \u001b[91m━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[91m╸\u001b[0m \u001b[32m2,115/2,120 \u001b[0m [ \u001b[33m0:00:02\u001b[0m < \u001b[36m0:00:01\u001b[0m , \u001b[31m7,715 it/s\u001b[0m ]t/s\u001b[0m ]\n",
            "\u001b[?25h\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m monophone - Iteration \u001b[1;36m25\u001b[0m of \u001b[1;36m40\u001b[0m                                        \n",
            "\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m Accumulating statistics\u001b[33m...\u001b[0m                                            \n",
            "\u001b[2K\u001b[35m  99%\u001b[0m \u001b[91m━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[91m╸\u001b[0m \u001b[32m2,093/2,120 \u001b[0m [ \u001b[33m0:00:02\u001b[0m < \u001b[36m0:00:01\u001b[0m , \u001b[31m7,504 it/s\u001b[0m ]t/s\u001b[0m ]\n",
            "\u001b[?25h\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m monophone - Iteration \u001b[1;36m26\u001b[0m of \u001b[1;36m40\u001b[0m                                        \n",
            "\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m Generating alignments\u001b[33m...\u001b[0m                                              \n",
            "\u001b[2K\u001b[35m  98%\u001b[0m \u001b[91m━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[91m╸\u001b[0m \u001b[32m2,084/2,120 \u001b[0m [ \u001b[33m0:00:05\u001b[0m < \u001b[36m0:00:01\u001b[0m , \u001b[31m706 it/s\u001b[0m ]t/s\u001b[0m ]\n",
            "\u001b[?25h\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m Accumulating statistics\u001b[33m...\u001b[0m                                            \n",
            "\u001b[2K\u001b[35m  92%\u001b[0m \u001b[91m━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[91m╸\u001b[0m\u001b[90m━━\u001b[0m \u001b[32m1,955/2,120 \u001b[0m [ \u001b[33m0:00:02\u001b[0m < \u001b[36m0:00:01\u001b[0m , \u001b[31m7,531 it/s\u001b[0m ]\n",
            "\u001b[?25h\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m monophone - Iteration \u001b[1;36m27\u001b[0m of \u001b[1;36m40\u001b[0m                                        \n",
            "\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m Accumulating statistics\u001b[33m...\u001b[0m                                            \n",
            "\u001b[2K\u001b[35m  97%\u001b[0m \u001b[91m━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m \u001b[32m2,058/2,120 \u001b[0m [ \u001b[33m0:00:02\u001b[0m < \u001b[36m0:00:01\u001b[0m , \u001b[31m7,486 it/s\u001b[0m ]t/s\u001b[0m ]\n",
            "\u001b[?25h\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m monophone - Iteration \u001b[1;36m28\u001b[0m of \u001b[1;36m40\u001b[0m                                        \n",
            "\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m Accumulating statistics\u001b[33m...\u001b[0m                                            \n",
            "\u001b[2K\u001b[35m  99%\u001b[0m \u001b[91m━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[91m╸\u001b[0m \u001b[32m2,102/2,120 \u001b[0m [ \u001b[33m0:00:02\u001b[0m < \u001b[36m0:00:01\u001b[0m , \u001b[31m7,464 it/s\u001b[0m ]t/s\u001b[0m ]\n",
            "\u001b[?25h\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m monophone - Iteration \u001b[1;36m29\u001b[0m of \u001b[1;36m40\u001b[0m                                        \n",
            "\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m Generating alignments\u001b[33m...\u001b[0m                                              \n",
            "\u001b[2K\u001b[35m 100%\u001b[0m \u001b[91m━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[91m╸\u001b[0m \u001b[32m2,114/2,120 \u001b[0m [ \u001b[33m0:00:05\u001b[0m < \u001b[36m0:00:01\u001b[0m , \u001b[31m729 it/s\u001b[0m ]t/s\u001b[0m ]\n",
            "\u001b[?25h\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m Accumulating statistics\u001b[33m...\u001b[0m                                            \n",
            "\u001b[2K\u001b[35m  98%\u001b[0m \u001b[91m━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m \u001b[32m2,073/2,120 \u001b[0m [ \u001b[33m0:00:02\u001b[0m < \u001b[36m0:00:01\u001b[0m , \u001b[31m5,990 it/s\u001b[0m ]t/s\u001b[0m ]\n",
            "\u001b[?25h\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m monophone - Iteration \u001b[1;36m30\u001b[0m of \u001b[1;36m40\u001b[0m                                        \n",
            "\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m Accumulating statistics\u001b[33m...\u001b[0m                                            \n",
            "\u001b[2K\u001b[35m 100%\u001b[0m \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m2,125/2,120 \u001b[0m [ \u001b[33m0:00:00\u001b[0m < \u001b[36m0:00:00\u001b[0m , \u001b[31m7,703 it/s\u001b[0m ]1m7,970 it/s\u001b[0m ]\n",
            "\u001b[?25h\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m monophone - Iteration \u001b[1;36m31\u001b[0m of \u001b[1;36m40\u001b[0m                                        \n",
            "\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m Accumulating statistics\u001b[33m...\u001b[0m                                            \n",
            "\u001b[2K\u001b[35m  98%\u001b[0m \u001b[91m━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m \u001b[32m2,070/2,120 \u001b[0m [ \u001b[33m0:00:02\u001b[0m < \u001b[36m0:00:01\u001b[0m , \u001b[31m7,188 it/s\u001b[0m ]t/s\u001b[0m ]\n",
            "\u001b[?25h\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m monophone - Iteration \u001b[1;36m32\u001b[0m of \u001b[1;36m40\u001b[0m                                        \n",
            "\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m Generating alignments\u001b[33m...\u001b[0m                                              \n",
            "\u001b[2K\u001b[35m  98%\u001b[0m \u001b[91m━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m \u001b[32m2,071/2,120 \u001b[0m [ \u001b[33m0:00:04\u001b[0m < \u001b[36m0:00:01\u001b[0m , \u001b[31m761 it/s\u001b[0m ]t/s\u001b[0m ]\n",
            "\u001b[?25h\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m Accumulating statistics\u001b[33m...\u001b[0m                                            \n",
            "\u001b[2K\u001b[35m  99%\u001b[0m \u001b[91m━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[91m╸\u001b[0m \u001b[32m2,103/2,120 \u001b[0m [ \u001b[33m0:00:02\u001b[0m < \u001b[36m0:00:01\u001b[0m , \u001b[31m7,644 it/s\u001b[0m ]t/s\u001b[0m ]\n",
            "\u001b[?25h\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m monophone - Iteration \u001b[1;36m33\u001b[0m of \u001b[1;36m40\u001b[0m                                        \n",
            "\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m Accumulating statistics\u001b[33m...\u001b[0m                                            \n",
            "\u001b[2K\u001b[35m 100%\u001b[0m \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m2,120/2,120 \u001b[0m [ \u001b[33m0:00:00\u001b[0m < \u001b[36m0:00:00\u001b[0m , \u001b[31m7,657 it/s\u001b[0m ]1m7,734 it/s\u001b[0m ]\n",
            "\u001b[?25h\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m monophone - Iteration \u001b[1;36m34\u001b[0m of \u001b[1;36m40\u001b[0m                                        \n",
            "\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m Accumulating statistics\u001b[33m...\u001b[0m                                            \n",
            "\u001b[2K\u001b[35m  99%\u001b[0m \u001b[91m━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[91m╸\u001b[0m \u001b[32m2,106/2,120 \u001b[0m [ \u001b[33m0:00:02\u001b[0m < \u001b[36m0:00:01\u001b[0m , \u001b[31m7,541 it/s\u001b[0m ]t/s\u001b[0m ]\n",
            "\u001b[?25h\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m monophone - Iteration \u001b[1;36m35\u001b[0m of \u001b[1;36m40\u001b[0m                                        \n",
            "\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m Generating alignments\u001b[33m...\u001b[0m                                              \n",
            "\u001b[2K\u001b[35m  98%\u001b[0m \u001b[91m━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m \u001b[32m2,068/2,120 \u001b[0m [ \u001b[33m0:00:04\u001b[0m < \u001b[36m0:00:01\u001b[0m , \u001b[31m799 it/s\u001b[0m ]t/s\u001b[0m ]\n",
            "\u001b[?25h\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m Accumulating statistics\u001b[33m...\u001b[0m                                            \n",
            "\u001b[2K\u001b[35m 100%\u001b[0m \u001b[91m━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[91m╸\u001b[0m \u001b[32m2,117/2,120 \u001b[0m [ \u001b[33m0:00:02\u001b[0m < \u001b[36m0:00:01\u001b[0m , \u001b[31m7,699 it/s\u001b[0m ]t/s\u001b[0m ]\n",
            "\u001b[?25h\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m monophone - Iteration \u001b[1;36m36\u001b[0m of \u001b[1;36m40\u001b[0m                                        \n",
            "\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m Accumulating statistics\u001b[33m...\u001b[0m                                            \n",
            "\u001b[2K\u001b[35m  94%\u001b[0m \u001b[91m━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━\u001b[0m \u001b[32m1,986/2,120 \u001b[0m [ \u001b[33m0:00:02\u001b[0m < \u001b[36m0:00:01\u001b[0m , \u001b[31m6,446 it/s\u001b[0m ]\n",
            "\u001b[?25h\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m monophone - Iteration \u001b[1;36m37\u001b[0m of \u001b[1;36m40\u001b[0m                                        \n",
            "\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m Accumulating statistics\u001b[33m...\u001b[0m                                            \n",
            "\u001b[2K\u001b[35m 100%\u001b[0m \u001b[91m━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[91m╸\u001b[0m \u001b[32m2,118/2,120 \u001b[0m [ \u001b[33m0:00:02\u001b[0m < \u001b[36m0:00:01\u001b[0m , \u001b[31m7,638 it/s\u001b[0m ]t/s\u001b[0m ]\n",
            "\u001b[?25h\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m monophone - Iteration \u001b[1;36m38\u001b[0m of \u001b[1;36m40\u001b[0m                                        \n",
            "\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m Generating alignments\u001b[33m...\u001b[0m                                              \n",
            "\u001b[2K\u001b[35m  98%\u001b[0m \u001b[91m━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[91m╸\u001b[0m \u001b[32m2,085/2,120 \u001b[0m [ \u001b[33m0:00:04\u001b[0m < \u001b[36m0:00:01\u001b[0m , \u001b[31m779 it/s\u001b[0m ]t/s\u001b[0m ]\n",
            "\u001b[?25h\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m Accumulating statistics\u001b[33m...\u001b[0m                                            \n",
            "\u001b[2K\u001b[35m  84%\u001b[0m \u001b[91m━━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[91m╸\u001b[0m\u001b[90m━━━━\u001b[0m \u001b[32m1,785/2,120 \u001b[0m [ \u001b[33m0:00:02\u001b[0m < \u001b[36m0:00:01\u001b[0m , \u001b[31m7,033 it/s\u001b[0m ]\n",
            "\u001b[?25h\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m monophone - Iteration \u001b[1;36m39\u001b[0m of \u001b[1;36m40\u001b[0m                                        \n",
            "\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m Accumulating statistics\u001b[33m...\u001b[0m                                            \n",
            "\u001b[2K\u001b[35m  98%\u001b[0m \u001b[91m━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m \u001b[32m2,074/2,120 \u001b[0m [ \u001b[33m0:00:02\u001b[0m < \u001b[36m0:00:01\u001b[0m , \u001b[31m7,540 it/s\u001b[0m ]t/s\u001b[0m ]\n",
            "\u001b[?25h\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m monophone - Iteration \u001b[1;36m40\u001b[0m of \u001b[1;36m40\u001b[0m                                        \n",
            "\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m Accumulating statistics\u001b[33m...\u001b[0m                                            \n",
            "\u001b[2K\u001b[35m  98%\u001b[0m \u001b[91m━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[91m╸\u001b[0m \u001b[32m2,081/2,120 \u001b[0m [ \u001b[33m0:00:02\u001b[0m < \u001b[36m0:00:01\u001b[0m , \u001b[31m7,586 it/s\u001b[0m ]t/s\u001b[0m ]\n",
            "\u001b[?25h\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m Training complete!                                                    \n",
            "\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m Compiling training graphs\u001b[33m...\u001b[0m                                          \n",
            "\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m Generating alignments\u001b[33m...\u001b[0m                                              \n",
            "\u001b[2K\u001b[35m  97%\u001b[0m \u001b[91m━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m \u001b[32m2,049/2,120 \u001b[0m [ \u001b[33m0:00:04\u001b[0m < \u001b[36m0:00:01\u001b[0m , \u001b[31m777 it/s\u001b[0m ]t/s\u001b[0m ]\n",
            "\u001b[?25h\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m Initializing training for triphone\u001b[33m...\u001b[0m                                 \n",
            "\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m Accumulating tree stats\u001b[33m...\u001b[0m                                            \n",
            "\u001b[2K\u001b[35m  98%\u001b[0m \u001b[91m━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[91m╸\u001b[0m \u001b[32m2,085/2,120 \u001b[0m [ \u001b[33m0:00:03\u001b[0m < \u001b[36m0:00:01\u001b[0m , \u001b[31m1,630 it/s\u001b[0m ]t/s\u001b[0m ]\n",
            "\u001b[?25h\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m Compiling training graphs\u001b[33m...\u001b[0m                                          \n",
            "\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m Converting alignments\u001b[33m...\u001b[0m                                              \n",
            "\u001b[2K\u001b[35m  53%\u001b[0m \u001b[91m━━━━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━━\u001b[0m \u001b[32m1,132/2,120 \u001b[0m [ \u001b[33m0:00:02\u001b[0m < \u001b[36m-:--:--\u001b[0m , \u001b[31m? it/s\u001b[0m ]\n",
            "\u001b[?25h\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m Initialization complete!                                              \n",
            "\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m triphone - Iteration \u001b[1;36m1\u001b[0m of \u001b[1;36m35\u001b[0m                                          \n",
            "\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m Accumulating statistics\u001b[33m...\u001b[0m                                            \n",
            "\u001b[2K\u001b[35m  94%\u001b[0m \u001b[91m━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[91m╸\u001b[0m\u001b[90m━\u001b[0m \u001b[32m2,001/2,120 \u001b[0m [ \u001b[33m0:00:02\u001b[0m < \u001b[36m0:00:01\u001b[0m , \u001b[31m7,288 it/s\u001b[0m ]\n",
            "\u001b[?25h\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m triphone - Iteration \u001b[1;36m2\u001b[0m of \u001b[1;36m35\u001b[0m                                          \n",
            "\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m Accumulating statistics\u001b[33m...\u001b[0m                                            \n",
            "\u001b[2K\u001b[35m  97%\u001b[0m \u001b[91m━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m \u001b[32m2,051/2,120 \u001b[0m [ \u001b[33m0:00:02\u001b[0m < \u001b[36m0:00:01\u001b[0m , \u001b[31m7,772 it/s\u001b[0m ]t/s\u001b[0m ]\n",
            "\u001b[?25h\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m triphone - Iteration \u001b[1;36m3\u001b[0m of \u001b[1;36m35\u001b[0m                                          \n",
            "\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m Accumulating statistics\u001b[33m...\u001b[0m                                            \n",
            "\u001b[2K\u001b[35m  99%\u001b[0m \u001b[91m━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[91m╸\u001b[0m \u001b[32m2,109/2,120 \u001b[0m [ \u001b[33m0:00:02\u001b[0m < \u001b[36m0:00:01\u001b[0m , \u001b[31m7,766 it/s\u001b[0m ]t/s\u001b[0m ]\n",
            "\u001b[?25h\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m triphone - Iteration \u001b[1;36m4\u001b[0m of \u001b[1;36m35\u001b[0m                                          \n",
            "\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m Accumulating statistics\u001b[33m...\u001b[0m                                            \n",
            "\u001b[2K\u001b[35m  93%\u001b[0m \u001b[91m━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━\u001b[0m \u001b[32m1,970/2,120 \u001b[0m [ \u001b[33m0:00:02\u001b[0m < \u001b[36m0:00:01\u001b[0m , \u001b[31m7,529 it/s\u001b[0m ]\n",
            "\u001b[?25h\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m triphone - Iteration \u001b[1;36m5\u001b[0m of \u001b[1;36m35\u001b[0m                                          \n",
            "\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m Accumulating statistics\u001b[33m...\u001b[0m                                            \n",
            "\u001b[2K\u001b[35m  90%\u001b[0m \u001b[91m━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━\u001b[0m \u001b[32m1,909/2,120 \u001b[0m [ \u001b[33m0:00:02\u001b[0m < \u001b[36m0:00:01\u001b[0m , \u001b[31m7,190 it/s\u001b[0m ]\n",
            "\u001b[?25h\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m triphone - Iteration \u001b[1;36m6\u001b[0m of \u001b[1;36m35\u001b[0m                                          \n",
            "\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m Accumulating statistics\u001b[33m...\u001b[0m                                            \n",
            "\u001b[2K\u001b[35m  98%\u001b[0m \u001b[91m━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[91m╸\u001b[0m \u001b[32m2,088/2,120 \u001b[0m [ \u001b[33m0:00:02\u001b[0m < \u001b[36m0:00:01\u001b[0m , \u001b[31m7,781 it/s\u001b[0m ]t/s\u001b[0m ]\n",
            "\u001b[?25h\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m triphone - Iteration \u001b[1;36m7\u001b[0m of \u001b[1;36m35\u001b[0m                                          \n",
            "\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m Accumulating statistics\u001b[33m...\u001b[0m                                            \n",
            "\u001b[2K\u001b[35m  97%\u001b[0m \u001b[91m━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m \u001b[32m2,055/2,120 \u001b[0m [ \u001b[33m0:00:02\u001b[0m < \u001b[36m0:00:01\u001b[0m , \u001b[31m7,867 it/s\u001b[0m ]t/s\u001b[0m ]\n",
            "\u001b[?25h\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m triphone - Iteration \u001b[1;36m8\u001b[0m of \u001b[1;36m35\u001b[0m                                          \n",
            "\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m Accumulating statistics\u001b[33m...\u001b[0m                                            \n",
            "\u001b[2K\u001b[35m  97%\u001b[0m \u001b[91m━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m \u001b[32m2,061/2,120 \u001b[0m [ \u001b[33m0:00:02\u001b[0m < \u001b[36m0:00:01\u001b[0m , \u001b[31m7,872 it/s\u001b[0m ]t/s\u001b[0m ]\n",
            "\u001b[?25h\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m triphone - Iteration \u001b[1;36m9\u001b[0m of \u001b[1;36m35\u001b[0m                                          \n",
            "\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m Accumulating statistics\u001b[33m...\u001b[0m                                            \n",
            "\u001b[2K\u001b[35m  80%\u001b[0m \u001b[91m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[91m╸\u001b[0m\u001b[90m━━━━━\u001b[0m \u001b[32m1,698/2,120 \u001b[0m [ \u001b[33m0:00:02\u001b[0m < \u001b[36m0:00:01\u001b[0m , \u001b[31m7,345 it/s\u001b[0m ]\n",
            "\u001b[?25h\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m triphone - Iteration \u001b[1;36m10\u001b[0m of \u001b[1;36m35\u001b[0m                                         \n",
            "\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m Generating alignments\u001b[33m...\u001b[0m                                              \n",
            "\u001b[2K\u001b[35m  97%\u001b[0m \u001b[91m━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m \u001b[32m2,059/2,120 \u001b[0m [ \u001b[33m0:00:04\u001b[0m < \u001b[36m0:00:01\u001b[0m , \u001b[31m837 it/s\u001b[0m ]t/s\u001b[0m ]\n",
            "\u001b[?25h\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m Accumulating statistics\u001b[33m...\u001b[0m                                            \n",
            "\u001b[2K\u001b[35m  90%\u001b[0m \u001b[91m━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━\u001b[0m \u001b[32m1,916/2,120 \u001b[0m [ \u001b[33m0:00:02\u001b[0m < \u001b[36m0:00:01\u001b[0m , \u001b[31m7,469 it/s\u001b[0m ]\n",
            "\u001b[?25h\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m triphone - Iteration \u001b[1;36m11\u001b[0m of \u001b[1;36m35\u001b[0m                                         \n",
            "\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m Accumulating statistics\u001b[33m...\u001b[0m                                            \n",
            "\u001b[2K\u001b[35m  90%\u001b[0m \u001b[91m━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━\u001b[0m \u001b[32m1,908/2,120 \u001b[0m [ \u001b[33m0:00:02\u001b[0m < \u001b[36m0:00:01\u001b[0m , \u001b[31m7,667 it/s\u001b[0m ]\n",
            "\u001b[?25h\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m triphone - Iteration \u001b[1;36m12\u001b[0m of \u001b[1;36m35\u001b[0m                                         \n",
            "\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m Accumulating statistics\u001b[33m...\u001b[0m                                            \n",
            "\u001b[2K\u001b[35m  89%\u001b[0m \u001b[91m━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━\u001b[0m \u001b[32m1,884/2,120 \u001b[0m [ \u001b[33m0:00:02\u001b[0m < \u001b[36m0:00:01\u001b[0m , \u001b[31m7,545 it/s\u001b[0m ]\n",
            "\u001b[?25h\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m triphone - Iteration \u001b[1;36m13\u001b[0m of \u001b[1;36m35\u001b[0m                                         \n",
            "\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m Accumulating statistics\u001b[33m...\u001b[0m                                            \n",
            "\u001b[2K\u001b[35m  89%\u001b[0m \u001b[91m━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━\u001b[0m \u001b[32m1,877/2,120 \u001b[0m [ \u001b[33m0:00:02\u001b[0m < \u001b[36m0:00:01\u001b[0m , \u001b[31m7,437 it/s\u001b[0m ]\n",
            "\u001b[?25h\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m triphone - Iteration \u001b[1;36m14\u001b[0m of \u001b[1;36m35\u001b[0m                                         \n",
            "\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m Accumulating statistics\u001b[33m...\u001b[0m                                            \n",
            "\u001b[2K\u001b[35m  88%\u001b[0m \u001b[91m━━━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[91m╸\u001b[0m\u001b[90m━━━\u001b[0m \u001b[32m1,856/2,120 \u001b[0m [ \u001b[33m0:00:02\u001b[0m < \u001b[36m0:00:01\u001b[0m , \u001b[31m7,373 it/s\u001b[0m ]\n",
            "\u001b[?25h\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m triphone - Iteration \u001b[1;36m15\u001b[0m of \u001b[1;36m35\u001b[0m                                         \n",
            "\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m Accumulating statistics\u001b[33m...\u001b[0m                                            \n",
            "\u001b[2K\u001b[35m  87%\u001b[0m \u001b[91m━━━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[91m╸\u001b[0m\u001b[90m━━━\u001b[0m \u001b[32m1,851/2,120 \u001b[0m [ \u001b[33m0:00:02\u001b[0m < \u001b[36m0:00:01\u001b[0m , \u001b[31m7,289 it/s\u001b[0m ]\n",
            "\u001b[?25h\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m triphone - Iteration \u001b[1;36m16\u001b[0m of \u001b[1;36m35\u001b[0m                                         \n",
            "\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m Accumulating statistics\u001b[33m...\u001b[0m                                            \n",
            "\u001b[2K\u001b[35m  83%\u001b[0m \u001b[91m━━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[91m╸\u001b[0m\u001b[90m━━━━\u001b[0m \u001b[32m1,769/2,120 \u001b[0m [ \u001b[33m0:00:02\u001b[0m < \u001b[36m0:00:01\u001b[0m , \u001b[31m7,055 it/s\u001b[0m ]\n",
            "\u001b[?25h\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m triphone - Iteration \u001b[1;36m17\u001b[0m of \u001b[1;36m35\u001b[0m                                         \n",
            "\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m Accumulating statistics\u001b[33m...\u001b[0m                                            \n",
            "\u001b[2K\u001b[35m  85%\u001b[0m \u001b[91m━━━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━\u001b[0m \u001b[32m1,803/2,120 \u001b[0m [ \u001b[33m0:00:02\u001b[0m < \u001b[36m0:00:01\u001b[0m , \u001b[31m7,294 it/s\u001b[0m ]\n",
            "\u001b[?25h\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m triphone - Iteration \u001b[1;36m18\u001b[0m of \u001b[1;36m35\u001b[0m                                         \n",
            "\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m Accumulating statistics\u001b[33m...\u001b[0m                                            \n",
            "\u001b[2K\u001b[35m  76%\u001b[0m \u001b[91m━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[91m╸\u001b[0m\u001b[90m━━━━━━\u001b[0m \u001b[32m1,619/2,120 \u001b[0m [ \u001b[33m0:00:02\u001b[0m < \u001b[36m0:00:01\u001b[0m , \u001b[31m6,428 it/s\u001b[0m ]\n",
            "\u001b[?25h\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m triphone - Iteration \u001b[1;36m19\u001b[0m of \u001b[1;36m35\u001b[0m                                         \n",
            "\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m Accumulating statistics\u001b[33m...\u001b[0m                                            \n",
            "\u001b[2K\u001b[35m  85%\u001b[0m \u001b[91m━━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[91m╸\u001b[0m\u001b[90m━━━━\u001b[0m \u001b[32m1,793/2,120 \u001b[0m [ \u001b[33m0:00:02\u001b[0m < \u001b[36m0:00:01\u001b[0m , \u001b[31m7,328 it/s\u001b[0m ]\n",
            "\u001b[?25h\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m triphone - Iteration \u001b[1;36m20\u001b[0m of \u001b[1;36m35\u001b[0m                                         \n",
            "\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m Generating alignments\u001b[33m...\u001b[0m                                              \n",
            "\u001b[2K\u001b[35m  97%\u001b[0m \u001b[91m━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m \u001b[32m2,063/2,120 \u001b[0m [ \u001b[33m0:00:04\u001b[0m < \u001b[36m0:00:01\u001b[0m , \u001b[31m845 it/s\u001b[0m ]t/s\u001b[0m ]\n",
            "\u001b[?25h\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m Accumulating statistics\u001b[33m...\u001b[0m                                            \n",
            "\u001b[2K\u001b[35m  90%\u001b[0m \u001b[91m━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━\u001b[0m \u001b[32m1,914/2,120 \u001b[0m [ \u001b[33m0:00:02\u001b[0m < \u001b[36m0:00:01\u001b[0m , \u001b[31m5,287 it/s\u001b[0m ]\n",
            "\u001b[?25h\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m triphone - Iteration \u001b[1;36m21\u001b[0m of \u001b[1;36m35\u001b[0m                                         \n",
            "\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m Accumulating statistics\u001b[33m...\u001b[0m                                            \n",
            "\u001b[2K\u001b[35m  81%\u001b[0m \u001b[91m━━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━\u001b[0m \u001b[32m1,722/2,120 \u001b[0m [ \u001b[33m0:00:02\u001b[0m < \u001b[36m0:00:01\u001b[0m , \u001b[31m6,994 it/s\u001b[0m ]\n",
            "\u001b[?25h\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m triphone - Iteration \u001b[1;36m22\u001b[0m of \u001b[1;36m35\u001b[0m                                         \n",
            "\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m Accumulating statistics\u001b[33m...\u001b[0m                                            \n",
            "\u001b[2K\u001b[35m  83%\u001b[0m \u001b[91m━━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━\u001b[0m \u001b[32m1,752/2,120 \u001b[0m [ \u001b[33m0:00:02\u001b[0m < \u001b[36m0:00:01\u001b[0m , \u001b[31m7,178 it/s\u001b[0m ]\n",
            "\u001b[?25h\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m triphone - Iteration \u001b[1;36m23\u001b[0m of \u001b[1;36m35\u001b[0m                                         \n",
            "\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m Accumulating statistics\u001b[33m...\u001b[0m                                            \n",
            "\u001b[2K\u001b[35m  77%\u001b[0m \u001b[91m━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[91m╸\u001b[0m\u001b[90m━━━━━━\u001b[0m \u001b[32m1,630/2,120 \u001b[0m [ \u001b[33m0:00:02\u001b[0m < \u001b[36m0:00:01\u001b[0m , \u001b[31m6,746 it/s\u001b[0m ]\n",
            "\u001b[?25h\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m triphone - Iteration \u001b[1;36m24\u001b[0m of \u001b[1;36m35\u001b[0m                                         \n",
            "\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m Accumulating statistics\u001b[33m...\u001b[0m                                            \n",
            "\u001b[2K\u001b[35m  76%\u001b[0m \u001b[91m━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[91m╸\u001b[0m\u001b[90m━━━━━━\u001b[0m \u001b[32m1,614/2,120 \u001b[0m [ \u001b[33m0:00:02\u001b[0m < \u001b[36m0:00:01\u001b[0m , \u001b[31m6,700 it/s\u001b[0m ]\n",
            "\u001b[?25h\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m triphone - Iteration \u001b[1;36m25\u001b[0m of \u001b[1;36m35\u001b[0m                                         \n",
            "\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m Accumulating statistics\u001b[33m...\u001b[0m                                            \n",
            "\u001b[2K\u001b[35m  74%\u001b[0m \u001b[91m━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━\u001b[0m \u001b[32m1,576/2,120 \u001b[0m [ \u001b[33m0:00:02\u001b[0m < \u001b[36m0:00:01\u001b[0m , \u001b[31m6,575 it/s\u001b[0m ]\n",
            "\u001b[?25h\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m triphone - Iteration \u001b[1;36m26\u001b[0m of \u001b[1;36m35\u001b[0m                                         \n",
            "\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m Accumulating statistics\u001b[33m...\u001b[0m                                            \n",
            "\u001b[2K\u001b[35m  77%\u001b[0m \u001b[91m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━\u001b[0m \u001b[32m1,641/2,120 \u001b[0m [ \u001b[33m0:00:02\u001b[0m < \u001b[36m0:00:01\u001b[0m , \u001b[31m6,946 it/s\u001b[0m ]\n",
            "\u001b[?25h\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m triphone - Iteration \u001b[1;36m27\u001b[0m of \u001b[1;36m35\u001b[0m                                         \n",
            "\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m Accumulating statistics\u001b[33m...\u001b[0m                                            \n",
            "\u001b[2K\u001b[35m 100%\u001b[0m \u001b[91m━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[91m╸\u001b[0m \u001b[32m2,110/2,120 \u001b[0m [ \u001b[33m0:00:02\u001b[0m < \u001b[36m0:00:01\u001b[0m , \u001b[31m6,209 it/s\u001b[0m ]t/s\u001b[0m ]\n",
            "\u001b[?25h\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m triphone - Iteration \u001b[1;36m28\u001b[0m of \u001b[1;36m35\u001b[0m                                         \n",
            "\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m Accumulating statistics\u001b[33m...\u001b[0m                                            \n",
            "\u001b[2K\u001b[35m  74%\u001b[0m \u001b[91m━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━\u001b[0m \u001b[32m1,566/2,120 \u001b[0m [ \u001b[33m0:00:02\u001b[0m < \u001b[36m0:00:01\u001b[0m , \u001b[31m6,701 it/s\u001b[0m ]\n",
            "\u001b[?25h\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m triphone - Iteration \u001b[1;36m29\u001b[0m of \u001b[1;36m35\u001b[0m                                         \n",
            "\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m Accumulating statistics\u001b[33m...\u001b[0m                                            \n",
            "\u001b[2K\u001b[35m  94%\u001b[0m \u001b[91m━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[91m╸\u001b[0m\u001b[90m━\u001b[0m \u001b[32m1,998/2,120 \u001b[0m [ \u001b[33m0:00:02\u001b[0m < \u001b[36m0:00:01\u001b[0m , \u001b[31m5,573 it/s\u001b[0m ]\n",
            "\u001b[?25h\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m triphone - Iteration \u001b[1;36m30\u001b[0m of \u001b[1;36m35\u001b[0m                                         \n",
            "\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m Generating alignments\u001b[33m...\u001b[0m                                              \n",
            "\u001b[2K\u001b[35m  98%\u001b[0m \u001b[91m━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m \u001b[32m2,076/2,120 \u001b[0m [ \u001b[33m0:00:04\u001b[0m < \u001b[36m0:00:01\u001b[0m , \u001b[31m839 it/s\u001b[0m ]t/s\u001b[0m ]\n",
            "\u001b[?25h\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m Accumulating statistics\u001b[33m...\u001b[0m                                            \n",
            "\u001b[2K\u001b[35m  74%\u001b[0m \u001b[91m━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━\u001b[0m \u001b[32m1,577/2,120 \u001b[0m [ \u001b[33m0:00:02\u001b[0m < \u001b[36m0:00:01\u001b[0m , \u001b[31m6,716 it/s\u001b[0m ]\n",
            "\u001b[?25h\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m triphone - Iteration \u001b[1;36m31\u001b[0m of \u001b[1;36m35\u001b[0m                                         \n",
            "\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m Accumulating statistics\u001b[33m...\u001b[0m                                            \n",
            "\u001b[2K\u001b[35m  97%\u001b[0m \u001b[91m━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m \u001b[32m2,054/2,120 \u001b[0m [ \u001b[33m0:00:02\u001b[0m < \u001b[36m0:00:01\u001b[0m , \u001b[31m5,925 it/s\u001b[0m ]t/s\u001b[0m ]\n",
            "\u001b[?25h\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m triphone - Iteration \u001b[1;36m32\u001b[0m of \u001b[1;36m35\u001b[0m                                         \n",
            "\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m Accumulating statistics\u001b[33m...\u001b[0m                                            \n",
            "\u001b[2K\u001b[35m 100%\u001b[0m \u001b[91m━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[91m╸\u001b[0m \u001b[32m2,111/2,120 \u001b[0m [ \u001b[33m0:00:02\u001b[0m < \u001b[36m0:00:01\u001b[0m , \u001b[31m6,198 it/s\u001b[0m ]t/s\u001b[0m ]\n",
            "\u001b[?25h\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m triphone - Iteration \u001b[1;36m33\u001b[0m of \u001b[1;36m35\u001b[0m                                         \n",
            "\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m Accumulating statistics\u001b[33m...\u001b[0m                                            \n",
            "\u001b[2K\u001b[35m  75%\u001b[0m \u001b[91m━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[91m╸\u001b[0m\u001b[90m━━━━━━\u001b[0m \u001b[32m1,590/2,120 \u001b[0m [ \u001b[33m0:00:02\u001b[0m < \u001b[36m0:00:01\u001b[0m , \u001b[31m6,663 it/s\u001b[0m ]\n",
            "\u001b[?25h\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m triphone - Iteration \u001b[1;36m34\u001b[0m of \u001b[1;36m35\u001b[0m                                         \n",
            "\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m Accumulating statistics\u001b[33m...\u001b[0m                                            \n",
            "\u001b[2K\u001b[35m  75%\u001b[0m \u001b[91m━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[91m╸\u001b[0m\u001b[90m━━━━━━\u001b[0m \u001b[32m1,598/2,120 \u001b[0m [ \u001b[33m0:00:02\u001b[0m < \u001b[36m0:00:01\u001b[0m , \u001b[31m6,598 it/s\u001b[0m ]\n",
            "\u001b[?25h\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m triphone - Iteration \u001b[1;36m35\u001b[0m of \u001b[1;36m35\u001b[0m                                         \n",
            "\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m Accumulating statistics\u001b[33m...\u001b[0m                                            \n",
            "\u001b[2K\u001b[35m  96%\u001b[0m \u001b[91m━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[91m╸\u001b[0m\u001b[90m━\u001b[0m \u001b[32m2,030/2,120 \u001b[0m [ \u001b[33m0:00:02\u001b[0m < \u001b[36m0:00:01\u001b[0m , \u001b[31m5,897 it/s\u001b[0m ]\n",
            "\u001b[?25h\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m Training complete!                                                    \n",
            "\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m Compiling training graphs\u001b[33m...\u001b[0m                                          \n",
            "\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m Generating alignments\u001b[33m...\u001b[0m                                              \n",
            "\u001b[2K\u001b[35m  97%\u001b[0m \u001b[91m━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m \u001b[32m2,062/2,120 \u001b[0m [ \u001b[33m0:00:04\u001b[0m < \u001b[36m0:00:01\u001b[0m , \u001b[31m843 it/s\u001b[0m ]t/s\u001b[0m ]\n",
            "\u001b[?25h\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m Initializing training for lda\u001b[33m...\u001b[0m                                      \n",
            "\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m Calculating initial LDA stats\u001b[33m...\u001b[0m                                      \n",
            "\u001b[2K\u001b[35m  98%\u001b[0m \u001b[91m━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[91m╸\u001b[0m \u001b[32m2,079/2,120 \u001b[0m [ \u001b[33m0:00:02\u001b[0m < \u001b[36m0:00:01\u001b[0m , \u001b[31m11,728 it/s\u001b[0m ]t/s\u001b[0m ]\n",
            "\u001b[?25h\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m Accumulating tree stats\u001b[33m...\u001b[0m                                            \n",
            "\u001b[2K\u001b[35m 100%\u001b[0m \u001b[91m━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[91m╸\u001b[0m \u001b[32m2,117/2,120 \u001b[0m [ \u001b[33m0:00:04\u001b[0m < \u001b[36m0:00:01\u001b[0m , \u001b[31m809 it/s\u001b[0m ]t/s\u001b[0m ]\n",
            "\u001b[?25h\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m Compiling training graphs\u001b[33m...\u001b[0m                                          \n",
            "\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m Converting alignments\u001b[33m...\u001b[0m                                              \n",
            "\u001b[2K\u001b[35m  36%\u001b[0m \u001b[91m━━━━━━━━━━━\u001b[0m\u001b[91m╸\u001b[0m\u001b[90m━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m773/2,120 \u001b[0m [ \u001b[33m0:00:02\u001b[0m < \u001b[36m-:--:--\u001b[0m , \u001b[31m? it/s\u001b[0m ]\n",
            "\u001b[?25h\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m Initialization complete!                                              \n",
            "\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m lda - Iteration \u001b[1;36m1\u001b[0m of \u001b[1;36m35\u001b[0m                                               \n",
            "\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m Accumulating statistics\u001b[33m...\u001b[0m                                            \n",
            "\u001b[2K\u001b[35m 100%\u001b[0m \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m2,127/2,120 \u001b[0m [ \u001b[33m0:00:02\u001b[0m < \u001b[36m0:00:00\u001b[0m , \u001b[31m755 it/s\u001b[0m ]t/s\u001b[0m ]t/s\u001b[0m ]\n",
            "\u001b[?25h\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m lda - Iteration \u001b[1;36m2\u001b[0m of \u001b[1;36m35\u001b[0m                                               \n",
            "\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m Re-calculating LDA\u001b[33m...\u001b[0m                                                 \n",
            "\u001b[2K\u001b[35m  97%\u001b[0m \u001b[91m━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m \u001b[32m2,062/2,120 \u001b[0m [ \u001b[33m0:00:05\u001b[0m < \u001b[36m0:00:01\u001b[0m , \u001b[31m549 it/s\u001b[0m ]t/s\u001b[0m ]\n",
            "\u001b[?25h\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m Accumulating statistics\u001b[33m...\u001b[0m                                            \n",
            "\u001b[2K\u001b[35m  98%\u001b[0m \u001b[91m━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m \u001b[32m2,075/2,120 \u001b[0m [ \u001b[33m0:00:05\u001b[0m < \u001b[36m0:00:01\u001b[0m , \u001b[31m661 it/s\u001b[0m ]t/s\u001b[0m ]\n",
            "\u001b[?25h\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m lda - Iteration \u001b[1;36m3\u001b[0m of \u001b[1;36m35\u001b[0m                                               \n",
            "\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m Accumulating statistics\u001b[33m...\u001b[0m                                            \n",
            "\u001b[2K\u001b[35m  99%\u001b[0m \u001b[91m━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[91m╸\u001b[0m \u001b[32m2,101/2,120 \u001b[0m [ \u001b[33m0:00:05\u001b[0m < \u001b[36m0:00:01\u001b[0m , \u001b[31m620 it/s\u001b[0m ]t/s\u001b[0m ]\n",
            "\u001b[?25h\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m lda - Iteration \u001b[1;36m4\u001b[0m of \u001b[1;36m35\u001b[0m                                               \n",
            "\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m Re-calculating LDA\u001b[33m...\u001b[0m                                                 \n",
            "\u001b[2K\u001b[35m  97%\u001b[0m \u001b[91m━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m \u001b[32m2,059/2,120 \u001b[0m [ \u001b[33m0:00:04\u001b[0m < \u001b[36m0:00:01\u001b[0m , \u001b[31m893 it/s\u001b[0m ]t/s\u001b[0m ]\n",
            "\u001b[?25h\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m Accumulating statistics\u001b[33m...\u001b[0m                                            \n",
            "\u001b[2K\u001b[35m  97%\u001b[0m \u001b[91m━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m \u001b[32m2,053/2,120 \u001b[0m [ \u001b[33m0:00:05\u001b[0m < \u001b[36m0:00:01\u001b[0m , \u001b[31m703 it/s\u001b[0m ]t/s\u001b[0m ]\n",
            "\u001b[?25h\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m lda - Iteration \u001b[1;36m5\u001b[0m of \u001b[1;36m35\u001b[0m                                               \n",
            "\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m Accumulating statistics\u001b[33m...\u001b[0m                                            \n",
            "\u001b[2K\u001b[35m  98%\u001b[0m \u001b[91m━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m \u001b[32m2,081/2,120 \u001b[0m [ \u001b[33m0:00:04\u001b[0m < \u001b[36m0:00:01\u001b[0m , \u001b[31m732 it/s\u001b[0m ]t/s\u001b[0m ]\n",
            "\u001b[?25h\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m lda - Iteration \u001b[1;36m6\u001b[0m of \u001b[1;36m35\u001b[0m                                               \n",
            "\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m Re-calculating LDA\u001b[33m...\u001b[0m                                                 \n",
            "\u001b[2K\u001b[35m  98%\u001b[0m \u001b[91m━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m \u001b[32m2,075/2,120 \u001b[0m [ \u001b[33m0:00:06\u001b[0m < \u001b[36m0:00:01\u001b[0m , \u001b[31m506 it/s\u001b[0m ]t/s\u001b[0m ]\n",
            "\u001b[?25h\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m Accumulating statistics\u001b[33m...\u001b[0m                                            \n",
            "\u001b[2K\u001b[35m 100%\u001b[0m \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m2,127/2,120 \u001b[0m [ \u001b[33m0:00:03\u001b[0m < \u001b[36m0:00:00\u001b[0m , \u001b[31m687 it/s\u001b[0m ]t/s\u001b[0m ]t/s\u001b[0m ]\n",
            "\u001b[?25h\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m lda - Iteration \u001b[1;36m7\u001b[0m of \u001b[1;36m35\u001b[0m                                               \n",
            "\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m Accumulating statistics\u001b[33m...\u001b[0m                                            \n",
            "\u001b[2K\u001b[35m  99%\u001b[0m \u001b[91m━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[91m╸\u001b[0m \u001b[32m2,100/2,120 \u001b[0m [ \u001b[33m0:00:06\u001b[0m < \u001b[36m0:00:01\u001b[0m , \u001b[31m532 it/s\u001b[0m ]t/s\u001b[0m ]\n",
            "\u001b[?25h\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m lda - Iteration \u001b[1;36m8\u001b[0m of \u001b[1;36m35\u001b[0m                                               \n",
            "\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m Accumulating statistics\u001b[33m...\u001b[0m                                            \n",
            "\u001b[2K\u001b[35m 100%\u001b[0m \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m2,122/2,120 \u001b[0m [ \u001b[33m0:00:03\u001b[0m < \u001b[36m0:00:00\u001b[0m , \u001b[31m672 it/s\u001b[0m ]t/s\u001b[0m ]t/s\u001b[0m ]\n",
            "\u001b[?25h\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m lda - Iteration \u001b[1;36m9\u001b[0m of \u001b[1;36m35\u001b[0m                                               \n",
            "\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m Accumulating statistics\u001b[33m...\u001b[0m                                            \n",
            "\u001b[2K\u001b[35m 100%\u001b[0m \u001b[91m━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[91m╸\u001b[0m \u001b[32m2,119/2,120 \u001b[0m [ \u001b[33m0:00:05\u001b[0m < \u001b[36m0:00:01\u001b[0m , \u001b[31m701 it/s\u001b[0m ]t/s\u001b[0m ]\n",
            "\u001b[?25h\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m lda - Iteration \u001b[1;36m10\u001b[0m of \u001b[1;36m35\u001b[0m                                              \n",
            "\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m Generating alignments\u001b[33m...\u001b[0m                                              \n",
            "\u001b[2K\u001b[35m 100%\u001b[0m \u001b[91m━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[91m╸\u001b[0m \u001b[32m2,116/2,120 \u001b[0m [ \u001b[33m0:00:18\u001b[0m < \u001b[36m0:00:01\u001b[0m , \u001b[31m132 it/s\u001b[0m ]t/s\u001b[0m ]\n",
            "\u001b[?25h\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m Accumulating statistics\u001b[33m...\u001b[0m                                            \n",
            "\u001b[2K\u001b[35m  99%\u001b[0m \u001b[91m━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[91m╸\u001b[0m \u001b[32m2,102/2,120 \u001b[0m [ \u001b[33m0:00:05\u001b[0m < \u001b[36m0:00:01\u001b[0m , \u001b[31m708 it/s\u001b[0m ]t/s\u001b[0m ]\n",
            "\u001b[?25h\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m lda - Iteration \u001b[1;36m11\u001b[0m of \u001b[1;36m35\u001b[0m                                              \n",
            "\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m Accumulating statistics\u001b[33m...\u001b[0m                                            \n",
            "\u001b[2K\u001b[35m  99%\u001b[0m \u001b[91m━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[91m╸\u001b[0m \u001b[32m2,092/2,120 \u001b[0m [ \u001b[33m0:00:05\u001b[0m < \u001b[36m0:00:01\u001b[0m , \u001b[31m621 it/s\u001b[0m ]t/s\u001b[0m ]\n",
            "\u001b[?25h\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m lda - Iteration \u001b[1;36m12\u001b[0m of \u001b[1;36m35\u001b[0m                                              \n",
            "\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m Re-calculating LDA\u001b[33m...\u001b[0m                                                 \n",
            "\u001b[2K\u001b[35m  98%\u001b[0m \u001b[91m━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m \u001b[32m2,076/2,120 \u001b[0m [ \u001b[33m0:00:05\u001b[0m < \u001b[36m0:00:01\u001b[0m , \u001b[31m583 it/s\u001b[0m ]t/s\u001b[0m ]\n",
            "\u001b[?25h\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m Accumulating statistics\u001b[33m...\u001b[0m                                            \n",
            "\u001b[2K\u001b[35m  98%\u001b[0m \u001b[91m━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m \u001b[32m2,080/2,120 \u001b[0m [ \u001b[33m0:00:05\u001b[0m < \u001b[36m0:00:01\u001b[0m , \u001b[31m668 it/s\u001b[0m ]t/s\u001b[0m ]\n",
            "\u001b[?25h\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m lda - Iteration \u001b[1;36m13\u001b[0m of \u001b[1;36m35\u001b[0m                                              \n",
            "\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m Accumulating statistics\u001b[33m...\u001b[0m                                            \n",
            "\u001b[2K\u001b[35m  98%\u001b[0m \u001b[91m━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m \u001b[32m2,078/2,120 \u001b[0m [ \u001b[33m0:00:06\u001b[0m < \u001b[36m0:00:01\u001b[0m , \u001b[31m491 it/s\u001b[0m ]t/s\u001b[0m ]\n",
            "\u001b[?25h\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m lda - Iteration \u001b[1;36m14\u001b[0m of \u001b[1;36m35\u001b[0m                                              \n",
            "\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m Accumulating statistics\u001b[33m...\u001b[0m                                            \n",
            "\u001b[2K\u001b[35m 100%\u001b[0m \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m2,123/2,120 \u001b[0m [ \u001b[33m0:00:03\u001b[0m < \u001b[36m0:00:00\u001b[0m , \u001b[31m648 it/s\u001b[0m ]t/s\u001b[0m ]t/s\u001b[0m ]\n",
            "\u001b[?25h\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m lda - Iteration \u001b[1;36m15\u001b[0m of \u001b[1;36m35\u001b[0m                                              \n",
            "\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m Accumulating statistics\u001b[33m...\u001b[0m                                            \n",
            "\u001b[2K\u001b[35m  98%\u001b[0m \u001b[91m━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m \u001b[32m2,075/2,120 \u001b[0m [ \u001b[33m0:00:05\u001b[0m < \u001b[36m0:00:01\u001b[0m , \u001b[31m620 it/s\u001b[0m ]t/s\u001b[0m ]\n",
            "\u001b[?25h\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m lda - Iteration \u001b[1;36m16\u001b[0m of \u001b[1;36m35\u001b[0m                                              \n",
            "\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m Accumulating statistics\u001b[33m...\u001b[0m                                            \n",
            "\u001b[2K\u001b[35m 100%\u001b[0m \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m2,127/2,120 \u001b[0m [ \u001b[33m0:00:03\u001b[0m < \u001b[36m0:00:00\u001b[0m , \u001b[31m612 it/s\u001b[0m ]t/s\u001b[0m ]t/s\u001b[0m ]\n",
            "\u001b[?25h\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m lda - Iteration \u001b[1;36m17\u001b[0m of \u001b[1;36m35\u001b[0m                                              \n",
            "\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m Accumulating statistics\u001b[33m...\u001b[0m                                            \n",
            "\u001b[2K\u001b[35m  99%\u001b[0m \u001b[91m━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[91m╸\u001b[0m \u001b[32m2,090/2,120 \u001b[0m [ \u001b[33m0:00:05\u001b[0m < \u001b[36m0:00:01\u001b[0m , \u001b[31m677 it/s\u001b[0m ]t/s\u001b[0m ]\n",
            "\u001b[?25h\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m lda - Iteration \u001b[1;36m18\u001b[0m of \u001b[1;36m35\u001b[0m                                              \n",
            "\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m Accumulating statistics\u001b[33m...\u001b[0m                                            \n",
            "\u001b[2K\u001b[35m 100%\u001b[0m \u001b[91m━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[91m╸\u001b[0m \u001b[32m2,115/2,120 \u001b[0m [ \u001b[33m0:00:05\u001b[0m < \u001b[36m0:00:01\u001b[0m , \u001b[31m658 it/s\u001b[0m ]t/s\u001b[0m ]\n",
            "\u001b[?25h\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m lda - Iteration \u001b[1;36m19\u001b[0m of \u001b[1;36m35\u001b[0m                                              \n",
            "\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m Accumulating statistics\u001b[33m...\u001b[0m                                            \n",
            "\u001b[2K\u001b[35m 100%\u001b[0m \u001b[91m━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[91m╸\u001b[0m \u001b[32m2,110/2,120 \u001b[0m [ \u001b[33m0:00:05\u001b[0m < \u001b[36m0:00:01\u001b[0m , \u001b[31m560 it/s\u001b[0m ]t/s\u001b[0m ]\n",
            "\u001b[?25h\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m lda - Iteration \u001b[1;36m20\u001b[0m of \u001b[1;36m35\u001b[0m                                              \n",
            "\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m Generating alignments\u001b[33m...\u001b[0m                                              \n",
            "\u001b[2K\u001b[35m  99%\u001b[0m \u001b[91m━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[91m╸\u001b[0m \u001b[32m2,102/2,120 \u001b[0m [ \u001b[33m0:00:17\u001b[0m < \u001b[36m0:00:01\u001b[0m , \u001b[31m134 it/s\u001b[0m ]t/s\u001b[0m ]\n",
            "\u001b[?25h\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m Accumulating statistics\u001b[33m...\u001b[0m                                            \n",
            "\u001b[2K\u001b[35m 100%\u001b[0m \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m2,120/2,120 \u001b[0m [ \u001b[33m0:00:03\u001b[0m < \u001b[36m0:00:00\u001b[0m , \u001b[31m640 it/s\u001b[0m ]t/s\u001b[0m ]t/s\u001b[0m ]\n",
            "\u001b[?25h\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m lda - Iteration \u001b[1;36m21\u001b[0m of \u001b[1;36m35\u001b[0m                                              \n",
            "\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m Accumulating statistics\u001b[33m...\u001b[0m                                            \n",
            "\u001b[2K\u001b[35m  98%\u001b[0m \u001b[91m━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m \u001b[32m2,071/2,120 \u001b[0m [ \u001b[33m0:00:05\u001b[0m < \u001b[36m0:00:01\u001b[0m , \u001b[31m548 it/s\u001b[0m ]t/s\u001b[0m ]\n",
            "\u001b[?25h\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m lda - Iteration \u001b[1;36m22\u001b[0m of \u001b[1;36m35\u001b[0m                                              \n",
            "\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m Accumulating statistics\u001b[33m...\u001b[0m                                            \n",
            "\u001b[2K\u001b[35m 100%\u001b[0m \u001b[91m━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[91m╸\u001b[0m \u001b[32m2,111/2,120 \u001b[0m [ \u001b[33m0:00:06\u001b[0m < \u001b[36m0:00:01\u001b[0m , \u001b[31m463 it/s\u001b[0m ]t/s\u001b[0m ]\n",
            "\u001b[?25h\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m lda - Iteration \u001b[1;36m23\u001b[0m of \u001b[1;36m35\u001b[0m                                              \n",
            "\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m Accumulating statistics\u001b[33m...\u001b[0m                                            \n",
            "\u001b[2K\u001b[35m  98%\u001b[0m \u001b[91m━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m \u001b[32m2,081/2,120 \u001b[0m [ \u001b[33m0:00:05\u001b[0m < \u001b[36m0:00:01\u001b[0m , \u001b[31m620 it/s\u001b[0m ]t/s\u001b[0m ]\n",
            "\u001b[?25h\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m lda - Iteration \u001b[1;36m24\u001b[0m of \u001b[1;36m35\u001b[0m                                              \n",
            "\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m Accumulating statistics\u001b[33m...\u001b[0m                                            \n",
            "\u001b[2K\u001b[35m  99%\u001b[0m \u001b[91m━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[91m╸\u001b[0m \u001b[32m2,096/2,120 \u001b[0m [ \u001b[33m0:00:05\u001b[0m < \u001b[36m0:00:01\u001b[0m , \u001b[31m555 it/s\u001b[0m ]t/s\u001b[0m ]\n",
            "\u001b[?25h\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m lda - Iteration \u001b[1;36m25\u001b[0m of \u001b[1;36m35\u001b[0m                                              \n",
            "\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m Accumulating statistics\u001b[33m...\u001b[0m                                            \n",
            "\u001b[2K\u001b[35m  99%\u001b[0m \u001b[91m━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[91m╸\u001b[0m \u001b[32m2,091/2,120 \u001b[0m [ \u001b[33m0:00:05\u001b[0m < \u001b[36m0:00:01\u001b[0m , \u001b[31m579 it/s\u001b[0m ]t/s\u001b[0m ]\n",
            "\u001b[?25h\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m lda - Iteration \u001b[1;36m26\u001b[0m of \u001b[1;36m35\u001b[0m                                              \n",
            "\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m Accumulating statistics\u001b[33m...\u001b[0m                                            \n",
            "\u001b[2K\u001b[35m  98%\u001b[0m \u001b[91m━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m \u001b[32m2,071/2,120 \u001b[0m [ \u001b[33m0:00:05\u001b[0m < \u001b[36m0:00:01\u001b[0m , \u001b[31m546 it/s\u001b[0m ]t/s\u001b[0m ]\n",
            "\u001b[?25h\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m lda - Iteration \u001b[1;36m27\u001b[0m of \u001b[1;36m35\u001b[0m                                              \n",
            "\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m Accumulating statistics\u001b[33m...\u001b[0m                                            \n",
            "\u001b[2K\u001b[35m 100%\u001b[0m \u001b[91m━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[91m╸\u001b[0m \u001b[32m2,110/2,120 \u001b[0m [ \u001b[33m0:00:06\u001b[0m < \u001b[36m0:00:01\u001b[0m , \u001b[31m523 it/s\u001b[0m ]t/s\u001b[0m ]\n",
            "\u001b[?25h\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m lda - Iteration \u001b[1;36m28\u001b[0m of \u001b[1;36m35\u001b[0m                                              \n",
            "\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m Accumulating statistics\u001b[33m...\u001b[0m                                            \n",
            "\u001b[2K\u001b[35m  99%\u001b[0m \u001b[91m━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[91m╸\u001b[0m \u001b[32m2,092/2,120 \u001b[0m [ \u001b[33m0:00:06\u001b[0m < \u001b[36m0:00:01\u001b[0m , \u001b[31m454 it/s\u001b[0m ]t/s\u001b[0m ]\n",
            "\u001b[?25h\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m lda - Iteration \u001b[1;36m29\u001b[0m of \u001b[1;36m35\u001b[0m                                              \n",
            "\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m Accumulating statistics\u001b[33m...\u001b[0m                                            \n",
            "\u001b[2K\u001b[35m  98%\u001b[0m \u001b[91m━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m \u001b[32m2,082/2,120 \u001b[0m [ \u001b[33m0:00:05\u001b[0m < \u001b[36m0:00:01\u001b[0m , \u001b[31m580 it/s\u001b[0m ]t/s\u001b[0m ]\n",
            "\u001b[?25h\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m lda - Iteration \u001b[1;36m30\u001b[0m of \u001b[1;36m35\u001b[0m                                              \n",
            "\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m Generating alignments\u001b[33m...\u001b[0m                                              \n",
            "\u001b[2K\u001b[35m 100%\u001b[0m \u001b[91m━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[91m╸\u001b[0m \u001b[32m2,112/2,120 \u001b[0m [ \u001b[33m0:00:17\u001b[0m < \u001b[36m0:00:01\u001b[0m , \u001b[31m136 it/s\u001b[0m ]t/s\u001b[0m ]\n",
            "\u001b[?25h\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m Accumulating statistics\u001b[33m...\u001b[0m                                            \n",
            "\u001b[2K\u001b[35m  98%\u001b[0m \u001b[91m━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m \u001b[32m2,073/2,120 \u001b[0m [ \u001b[33m0:00:05\u001b[0m < \u001b[36m0:00:01\u001b[0m , \u001b[31m558 it/s\u001b[0m ]t/s\u001b[0m ]\n",
            "\u001b[?25h\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m lda - Iteration \u001b[1;36m31\u001b[0m of \u001b[1;36m35\u001b[0m                                              \n",
            "\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m Accumulating statistics\u001b[33m...\u001b[0m                                            \n",
            "\u001b[2K\u001b[35m  98%\u001b[0m \u001b[91m━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m \u001b[32m2,073/2,120 \u001b[0m [ \u001b[33m0:00:05\u001b[0m < \u001b[36m0:00:01\u001b[0m , \u001b[31m550 it/s\u001b[0m ]t/s\u001b[0m ]\n",
            "\u001b[?25h\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m lda - Iteration \u001b[1;36m32\u001b[0m of \u001b[1;36m35\u001b[0m                                              \n",
            "\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m Accumulating statistics\u001b[33m...\u001b[0m                                            \n",
            "\u001b[2K\u001b[35m  99%\u001b[0m \u001b[91m━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[91m╸\u001b[0m \u001b[32m2,095/2,120 \u001b[0m [ \u001b[33m0:00:06\u001b[0m < \u001b[36m0:00:01\u001b[0m , \u001b[31m515 it/s\u001b[0m ]t/s\u001b[0m ]\n",
            "\u001b[?25h\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m lda - Iteration \u001b[1;36m33\u001b[0m of \u001b[1;36m35\u001b[0m                                              \n",
            "\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m Accumulating statistics\u001b[33m...\u001b[0m                                            \n",
            "\u001b[2K\u001b[35m 100%\u001b[0m \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m2,127/2,120 \u001b[0m [ \u001b[33m0:00:03\u001b[0m < \u001b[36m0:00:00\u001b[0m , \u001b[31m602 it/s\u001b[0m ]t/s\u001b[0m ]t/s\u001b[0m ]\n",
            "\u001b[?25h\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m lda - Iteration \u001b[1;36m34\u001b[0m of \u001b[1;36m35\u001b[0m                                              \n",
            "\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m Accumulating statistics\u001b[33m...\u001b[0m                                            \n",
            "\u001b[2K\u001b[35m  99%\u001b[0m \u001b[91m━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[91m╸\u001b[0m \u001b[32m2,105/2,120 \u001b[0m [ \u001b[33m0:00:05\u001b[0m < \u001b[36m0:00:01\u001b[0m , \u001b[31m610 it/s\u001b[0m ]t/s\u001b[0m ]\n",
            "\u001b[?25h\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m lda - Iteration \u001b[1;36m35\u001b[0m of \u001b[1;36m35\u001b[0m                                              \n",
            "\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m Accumulating statistics\u001b[33m...\u001b[0m                                            \n",
            "\u001b[2K\u001b[35m 100%\u001b[0m \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m2,123/2,120 \u001b[0m [ \u001b[33m0:00:04\u001b[0m < \u001b[36m0:00:00\u001b[0m , \u001b[31m495 it/s\u001b[0m ]t/s\u001b[0m ]t/s\u001b[0m ]\n",
            "\u001b[?25h\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m Training complete!                                                    \n",
            "\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m Compiling training graphs\u001b[33m...\u001b[0m                                          \n",
            "\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m Generating alignments\u001b[33m...\u001b[0m                                              \n",
            "\u001b[2K\u001b[35m 100%\u001b[0m \u001b[91m━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[91m╸\u001b[0m \u001b[32m2,114/2,120 \u001b[0m [ \u001b[33m0:00:19\u001b[0m < \u001b[36m0:00:01\u001b[0m , \u001b[31m124 it/s\u001b[0m ]t/s\u001b[0m ]\n",
            "\u001b[?25h\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m Initializing training for sat\u001b[33m...\u001b[0m                                      \n",
            "\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m Calculating fMLLR for speaker adaptation\u001b[33m...\u001b[0m                           \n",
            "\u001b[2K\u001b[35m 100%\u001b[0m \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m9/1 \u001b[0m [ \u001b[33m0:00:07\u001b[0m < \u001b[36m0:00:00\u001b[0m , \u001b[31m8 it/s\u001b[0m ]\n",
            "\u001b[?25h\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m Accumulating tree stats\u001b[33m...\u001b[0m                                            \n",
            "\u001b[2K\u001b[35m  97%\u001b[0m \u001b[91m━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m \u001b[32m2,048/2,120 \u001b[0m [ \u001b[33m0:00:04\u001b[0m < \u001b[36m0:00:01\u001b[0m , \u001b[31m925 it/s\u001b[0m ]t/s\u001b[0m ]\n",
            "\u001b[?25h\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m Converting alignments\u001b[33m...\u001b[0m                                              \n",
            "\u001b[2K\u001b[35m  32%\u001b[0m \u001b[91m━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m689/2,120 \u001b[0m [ \u001b[33m0:00:02\u001b[0m < \u001b[36m-:--:--\u001b[0m , \u001b[31m? it/s\u001b[0m ]\n",
            "\u001b[?25h\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m Compiling training graphs\u001b[33m...\u001b[0m                                          \n",
            "\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m Initialization complete!                                              \n",
            "\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m sat - Iteration \u001b[1;36m1\u001b[0m of \u001b[1;36m35\u001b[0m                                               \n",
            "\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m Accumulating statistics\u001b[33m...\u001b[0m                                            \n",
            "\u001b[2K\u001b[35m  99%\u001b[0m \u001b[91m━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[91m╸\u001b[0m \u001b[32m2,089/2,120 \u001b[0m [ \u001b[33m0:00:04\u001b[0m < \u001b[36m0:00:01\u001b[0m , \u001b[31m795 it/s\u001b[0m ]t/s\u001b[0m ]\n",
            "\u001b[?25h\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m sat - Iteration \u001b[1;36m2\u001b[0m of \u001b[1;36m35\u001b[0m                                               \n",
            "\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m Calculating fMLLR for speaker adaptation\u001b[33m...\u001b[0m                           \n",
            "\u001b[2K\u001b[35m 100%\u001b[0m \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m10/1 \u001b[0m [ \u001b[33m0:00:06\u001b[0m < \u001b[36m0:00:00\u001b[0m , \u001b[31m11 it/s\u001b[0m ]\n",
            "\u001b[?25h\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m Accumulating statistics\u001b[33m...\u001b[0m                                            \n",
            "\u001b[2K\u001b[35m 100%\u001b[0m \u001b[91m━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[91m╸\u001b[0m \u001b[32m2,115/2,120 \u001b[0m [ \u001b[33m0:00:04\u001b[0m < \u001b[36m0:00:01\u001b[0m , \u001b[31m732 it/s\u001b[0m ]t/s\u001b[0m ]\n",
            "\u001b[?25h\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m sat - Iteration \u001b[1;36m3\u001b[0m of \u001b[1;36m35\u001b[0m                                               \n",
            "\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m Accumulating statistics\u001b[33m...\u001b[0m                                            \n",
            "\u001b[2K\u001b[35m 100%\u001b[0m \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m2,120/2,120 \u001b[0m [ \u001b[33m0:00:02\u001b[0m < \u001b[36m0:00:00\u001b[0m , \u001b[31m794 it/s\u001b[0m ][31m795 it/s\u001b[0m ]\n",
            "\u001b[?25h\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m sat - Iteration \u001b[1;36m4\u001b[0m of \u001b[1;36m35\u001b[0m                                               \n",
            "\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m Calculating fMLLR for speaker adaptation\u001b[33m...\u001b[0m                           \n",
            "\u001b[2K\u001b[35m 100%\u001b[0m \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m11/1 \u001b[0m [ \u001b[33m0:00:06\u001b[0m < \u001b[36m0:00:00\u001b[0m , \u001b[31m8 it/s\u001b[0m ]\n",
            "\u001b[?25h\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m Accumulating statistics\u001b[33m...\u001b[0m                                            \n",
            "\u001b[2K\u001b[35m 100%\u001b[0m \u001b[91m━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[91m╸\u001b[0m \u001b[32m2,112/2,120 \u001b[0m [ \u001b[33m0:00:05\u001b[0m < \u001b[36m0:00:01\u001b[0m , \u001b[31m649 it/s\u001b[0m ]t/s\u001b[0m ]\n",
            "\u001b[?25h\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m sat - Iteration \u001b[1;36m5\u001b[0m of \u001b[1;36m35\u001b[0m                                               \n",
            "\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m Accumulating statistics\u001b[33m...\u001b[0m                                            \n",
            "\u001b[2K\u001b[35m  96%\u001b[0m \u001b[91m━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[91m╸\u001b[0m\u001b[90m━\u001b[0m \u001b[32m2,031/2,120 \u001b[0m [ \u001b[33m0:00:04\u001b[0m < \u001b[36m0:00:01\u001b[0m , \u001b[31m896 it/s\u001b[0m ]\n",
            "\u001b[?25h\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m sat - Iteration \u001b[1;36m6\u001b[0m of \u001b[1;36m35\u001b[0m                                               \n",
            "\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m Calculating fMLLR for speaker adaptation\u001b[33m...\u001b[0m                           \n",
            "\u001b[2K\u001b[35m 100%\u001b[0m \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m12/1 \u001b[0m [ \u001b[33m0:00:06\u001b[0m < \u001b[36m0:00:00\u001b[0m , \u001b[31m8 it/s\u001b[0m ]\n",
            "\u001b[?25h\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m Accumulating statistics\u001b[33m...\u001b[0m                                            \n",
            "\u001b[2K\u001b[35m  97%\u001b[0m \u001b[91m━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m \u001b[32m2,064/2,120 \u001b[0m [ \u001b[33m0:00:04\u001b[0m < \u001b[36m0:00:01\u001b[0m , \u001b[31m737 it/s\u001b[0m ]t/s\u001b[0m ]\n",
            "\u001b[?25h\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m sat - Iteration \u001b[1;36m7\u001b[0m of \u001b[1;36m35\u001b[0m                                               \n",
            "\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m Accumulating statistics\u001b[33m...\u001b[0m                                            \n",
            "\u001b[2K\u001b[35m  97%\u001b[0m \u001b[91m━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m \u001b[32m2,062/2,120 \u001b[0m [ \u001b[33m0:00:05\u001b[0m < \u001b[36m0:00:01\u001b[0m , \u001b[31m707 it/s\u001b[0m ]t/s\u001b[0m ]\n",
            "\u001b[?25h\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m sat - Iteration \u001b[1;36m8\u001b[0m of \u001b[1;36m35\u001b[0m                                               \n",
            "\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m Accumulating statistics\u001b[33m...\u001b[0m                                            \n",
            "\u001b[2K\u001b[35m  99%\u001b[0m \u001b[91m━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[91m╸\u001b[0m \u001b[32m2,097/2,120 \u001b[0m [ \u001b[33m0:00:05\u001b[0m < \u001b[36m0:00:01\u001b[0m , \u001b[31m715 it/s\u001b[0m ]t/s\u001b[0m ]\n",
            "\u001b[?25h\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m sat - Iteration \u001b[1;36m9\u001b[0m of \u001b[1;36m35\u001b[0m                                               \n",
            "\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m Accumulating statistics\u001b[33m...\u001b[0m                                            \n",
            "\u001b[2K\u001b[35m  98%\u001b[0m \u001b[91m━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[91m╸\u001b[0m \u001b[32m2,086/2,120 \u001b[0m [ \u001b[33m0:00:04\u001b[0m < \u001b[36m0:00:01\u001b[0m , \u001b[31m845 it/s\u001b[0m ]t/s\u001b[0m ]\n",
            "\u001b[?25h\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m sat - Iteration \u001b[1;36m10\u001b[0m of \u001b[1;36m35\u001b[0m                                              \n",
            "\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m Generating alignments\u001b[33m...\u001b[0m                                              \n",
            "\u001b[2K\u001b[35m  99%\u001b[0m \u001b[91m━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[91m╸\u001b[0m \u001b[32m2,108/2,120 \u001b[0m [ \u001b[33m0:00:17\u001b[0m < \u001b[36m0:00:01\u001b[0m , \u001b[31m140 it/s\u001b[0m ]t/s\u001b[0m ]\n",
            "\u001b[?25h\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m Accumulating statistics\u001b[33m...\u001b[0m                                            \n",
            "\u001b[2K\u001b[35m  99%\u001b[0m \u001b[91m━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[91m╸\u001b[0m \u001b[32m2,098/2,120 \u001b[0m [ \u001b[33m0:00:05\u001b[0m < \u001b[36m0:00:01\u001b[0m , \u001b[31m688 it/s\u001b[0m ]t/s\u001b[0m ]\n",
            "\u001b[?25h\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m sat - Iteration \u001b[1;36m11\u001b[0m of \u001b[1;36m35\u001b[0m                                              \n",
            "\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m Accumulating statistics\u001b[33m...\u001b[0m                                            \n",
            "\u001b[2K\u001b[35m 100%\u001b[0m \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m2,128/2,120 \u001b[0m [ \u001b[33m0:00:02\u001b[0m < \u001b[36m0:00:00\u001b[0m , \u001b[31m810 it/s\u001b[0m ]t/s\u001b[0m ]t/s\u001b[0m ]\n",
            "\u001b[?25h\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m sat - Iteration \u001b[1;36m12\u001b[0m of \u001b[1;36m35\u001b[0m                                              \n",
            "\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m Calculating fMLLR for speaker adaptation\u001b[33m...\u001b[0m                           \n",
            "\u001b[2K\u001b[35m 100%\u001b[0m \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m11/1 \u001b[0m [ \u001b[33m0:00:06\u001b[0m < \u001b[36m0:00:00\u001b[0m , \u001b[31m8 it/s\u001b[0m ]\n",
            "\u001b[?25h\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m Accumulating statistics\u001b[33m...\u001b[0m                                            \n",
            "\u001b[2K\u001b[35m  98%\u001b[0m \u001b[91m━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m \u001b[32m2,080/2,120 \u001b[0m [ \u001b[33m0:00:05\u001b[0m < \u001b[36m0:00:01\u001b[0m , \u001b[31m689 it/s\u001b[0m ]t/s\u001b[0m ]\n",
            "\u001b[?25h\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m sat - Iteration \u001b[1;36m13\u001b[0m of \u001b[1;36m35\u001b[0m                                              \n",
            "\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m Accumulating statistics\u001b[33m...\u001b[0m                                            \n",
            "\u001b[2K\u001b[35m 100%\u001b[0m \u001b[91m━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[91m╸\u001b[0m \u001b[32m2,112/2,120 \u001b[0m [ \u001b[33m0:00:05\u001b[0m < \u001b[36m0:00:01\u001b[0m , \u001b[31m671 it/s\u001b[0m ]t/s\u001b[0m ]\n",
            "\u001b[?25h\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m sat - Iteration \u001b[1;36m14\u001b[0m of \u001b[1;36m35\u001b[0m                                              \n",
            "\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m Accumulating statistics\u001b[33m...\u001b[0m                                            \n",
            "\u001b[2K\u001b[35m  99%\u001b[0m \u001b[91m━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[91m╸\u001b[0m \u001b[32m2,095/2,120 \u001b[0m [ \u001b[33m0:00:05\u001b[0m < \u001b[36m0:00:01\u001b[0m , \u001b[31m698 it/s\u001b[0m ]t/s\u001b[0m ]\n",
            "\u001b[?25h\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m sat - Iteration \u001b[1;36m15\u001b[0m of \u001b[1;36m35\u001b[0m                                              \n",
            "\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m Accumulating statistics\u001b[33m...\u001b[0m                                            \n",
            "\u001b[2K\u001b[35m 100%\u001b[0m \u001b[91m━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[91m╸\u001b[0m \u001b[32m2,119/2,120 \u001b[0m [ \u001b[33m0:00:05\u001b[0m < \u001b[36m0:00:01\u001b[0m , \u001b[31m672 it/s\u001b[0m ]t/s\u001b[0m ]\n",
            "\u001b[?25h\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m sat - Iteration \u001b[1;36m16\u001b[0m of \u001b[1;36m35\u001b[0m                                              \n",
            "\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m Accumulating statistics\u001b[33m...\u001b[0m                                            \n",
            "\u001b[2K\u001b[35m 100%\u001b[0m \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m2,127/2,120 \u001b[0m [ \u001b[33m0:00:03\u001b[0m < \u001b[36m0:00:00\u001b[0m , \u001b[31m682 it/s\u001b[0m ]t/s\u001b[0m ]t/s\u001b[0m ]\n",
            "\u001b[?25h\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m sat - Iteration \u001b[1;36m17\u001b[0m of \u001b[1;36m35\u001b[0m                                              \n",
            "\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m Accumulating statistics\u001b[33m...\u001b[0m                                            \n",
            "\u001b[2K\u001b[35m  98%\u001b[0m \u001b[91m━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m \u001b[32m2,074/2,120 \u001b[0m [ \u001b[33m0:00:04\u001b[0m < \u001b[36m0:00:01\u001b[0m , \u001b[31m890 it/s\u001b[0m ]t/s\u001b[0m ]\n",
            "\u001b[?25h\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m sat - Iteration \u001b[1;36m18\u001b[0m of \u001b[1;36m35\u001b[0m                                              \n",
            "\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m Accumulating statistics\u001b[33m...\u001b[0m                                            \n",
            "\u001b[2K\u001b[35m 100%\u001b[0m \u001b[91m━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[91m╸\u001b[0m \u001b[32m2,113/2,120 \u001b[0m [ \u001b[33m0:00:04\u001b[0m < \u001b[36m0:00:01\u001b[0m , \u001b[31m801 it/s\u001b[0m ]t/s\u001b[0m ]\n",
            "\u001b[?25h\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m sat - Iteration \u001b[1;36m19\u001b[0m of \u001b[1;36m35\u001b[0m                                              \n",
            "\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m Accumulating statistics\u001b[33m...\u001b[0m                                            \n",
            "\u001b[2K\u001b[35m  99%\u001b[0m \u001b[91m━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[91m╸\u001b[0m \u001b[32m2,108/2,120 \u001b[0m [ \u001b[33m0:00:05\u001b[0m < \u001b[36m0:00:01\u001b[0m , \u001b[31m704 it/s\u001b[0m ]t/s\u001b[0m ]\n",
            "\u001b[?25h\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m sat - Iteration \u001b[1;36m20\u001b[0m of \u001b[1;36m35\u001b[0m                                              \n",
            "\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m Generating alignments\u001b[33m...\u001b[0m                                              \n",
            "\u001b[2K\u001b[35m  99%\u001b[0m \u001b[91m━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[91m╸\u001b[0m \u001b[32m2,104/2,120 \u001b[0m [ \u001b[33m0:00:15\u001b[0m < \u001b[36m0:00:01\u001b[0m , \u001b[31m152 it/s\u001b[0m ]t/s\u001b[0m ]\n",
            "\u001b[?25h\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m Accumulating statistics\u001b[33m...\u001b[0m                                            \n",
            "\u001b[2K\u001b[35m  98%\u001b[0m \u001b[91m━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m \u001b[32m2,075/2,120 \u001b[0m [ \u001b[33m0:00:05\u001b[0m < \u001b[36m0:00:01\u001b[0m , \u001b[31m697 it/s\u001b[0m ]t/s\u001b[0m ]\n",
            "\u001b[?25h\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m sat - Iteration \u001b[1;36m21\u001b[0m of \u001b[1;36m35\u001b[0m                                              \n",
            "\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m Accumulating statistics\u001b[33m...\u001b[0m                                            \n",
            "\u001b[2K\u001b[35m  99%\u001b[0m \u001b[91m━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[91m╸\u001b[0m \u001b[32m2,100/2,120 \u001b[0m [ \u001b[33m0:00:05\u001b[0m < \u001b[36m0:00:01\u001b[0m , \u001b[31m658 it/s\u001b[0m ]t/s\u001b[0m ]\n",
            "\u001b[?25h\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m sat - Iteration \u001b[1;36m22\u001b[0m of \u001b[1;36m35\u001b[0m                                              \n",
            "\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m Accumulating statistics\u001b[33m...\u001b[0m                                            \n",
            "\u001b[2K\u001b[35m  99%\u001b[0m \u001b[91m━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[91m╸\u001b[0m \u001b[32m2,093/2,120 \u001b[0m [ \u001b[33m0:00:05\u001b[0m < \u001b[36m0:00:01\u001b[0m , \u001b[31m664 it/s\u001b[0m ]t/s\u001b[0m ]\n",
            "\u001b[?25h\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m sat - Iteration \u001b[1;36m23\u001b[0m of \u001b[1;36m35\u001b[0m                                              \n",
            "\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m Accumulating statistics\u001b[33m...\u001b[0m                                            \n",
            "\u001b[2K\u001b[35m  98%\u001b[0m \u001b[91m━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[91m╸\u001b[0m \u001b[32m2,087/2,120 \u001b[0m [ \u001b[33m0:00:05\u001b[0m < \u001b[36m0:00:01\u001b[0m , \u001b[31m653 it/s\u001b[0m ]t/s\u001b[0m ]\n",
            "\u001b[?25h\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m sat - Iteration \u001b[1;36m24\u001b[0m of \u001b[1;36m35\u001b[0m                                              \n",
            "\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m Accumulating statistics\u001b[33m...\u001b[0m                                            \n",
            "\u001b[2K\u001b[35m 100%\u001b[0m \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m2,122/2,120 \u001b[0m [ \u001b[33m0:00:03\u001b[0m < \u001b[36m0:00:00\u001b[0m , \u001b[31m622 it/s\u001b[0m ]t/s\u001b[0m ]t/s\u001b[0m ]\n",
            "\u001b[?25h\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m sat - Iteration \u001b[1;36m25\u001b[0m of \u001b[1;36m35\u001b[0m                                              \n",
            "\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m Accumulating statistics\u001b[33m...\u001b[0m                                            \n",
            "\u001b[2K\u001b[35m  98%\u001b[0m \u001b[91m━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m \u001b[32m2,078/2,120 \u001b[0m [ \u001b[33m0:00:05\u001b[0m < \u001b[36m0:00:01\u001b[0m , \u001b[31m734 it/s\u001b[0m ]t/s\u001b[0m ]\n",
            "\u001b[?25h\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m sat - Iteration \u001b[1;36m26\u001b[0m of \u001b[1;36m35\u001b[0m                                              \n",
            "\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m Accumulating statistics\u001b[33m...\u001b[0m                                            \n",
            "\u001b[2K\u001b[35m  99%\u001b[0m \u001b[91m━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[91m╸\u001b[0m \u001b[32m2,108/2,120 \u001b[0m [ \u001b[33m0:00:05\u001b[0m < \u001b[36m0:00:01\u001b[0m , \u001b[31m609 it/s\u001b[0m ]t/s\u001b[0m ]\n",
            "\u001b[?25h\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m sat - Iteration \u001b[1;36m27\u001b[0m of \u001b[1;36m35\u001b[0m                                              \n",
            "\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m Accumulating statistics\u001b[33m...\u001b[0m                                            \n",
            "\u001b[2K\u001b[35m  99%\u001b[0m \u001b[91m━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[91m╸\u001b[0m \u001b[32m2,105/2,120 \u001b[0m [ \u001b[33m0:00:05\u001b[0m < \u001b[36m0:00:01\u001b[0m , \u001b[31m691 it/s\u001b[0m ]t/s\u001b[0m ]\n",
            "\u001b[?25h\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m sat - Iteration \u001b[1;36m28\u001b[0m of \u001b[1;36m35\u001b[0m                                              \n",
            "\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m Accumulating statistics\u001b[33m...\u001b[0m                                            \n",
            "\u001b[2K\u001b[35m 100%\u001b[0m \u001b[91m━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[91m╸\u001b[0m \u001b[32m2,111/2,120 \u001b[0m [ \u001b[33m0:00:05\u001b[0m < \u001b[36m0:00:01\u001b[0m , \u001b[31m684 it/s\u001b[0m ]t/s\u001b[0m ]\n",
            "\u001b[?25h\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m sat - Iteration \u001b[1;36m29\u001b[0m of \u001b[1;36m35\u001b[0m                                              \n",
            "\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m Accumulating statistics\u001b[33m...\u001b[0m                                            \n",
            "\u001b[2K\u001b[35m  99%\u001b[0m \u001b[91m━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[91m╸\u001b[0m \u001b[32m2,106/2,120 \u001b[0m [ \u001b[33m0:00:05\u001b[0m < \u001b[36m0:00:01\u001b[0m , \u001b[31m628 it/s\u001b[0m ]t/s\u001b[0m ]\n",
            "\u001b[?25h\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m sat - Iteration \u001b[1;36m30\u001b[0m of \u001b[1;36m35\u001b[0m                                              \n",
            "\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m Generating alignments\u001b[33m...\u001b[0m                                              \n",
            "\u001b[2K\u001b[35m  99%\u001b[0m \u001b[91m━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[91m╸\u001b[0m \u001b[32m2,105/2,120 \u001b[0m [ \u001b[33m0:00:16\u001b[0m < \u001b[36m0:00:01\u001b[0m , \u001b[31m143 it/s\u001b[0m ]t/s\u001b[0m ]\n",
            "\u001b[?25h\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m Accumulating statistics\u001b[33m...\u001b[0m                                            \n",
            "\u001b[2K\u001b[35m  98%\u001b[0m \u001b[91m━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m \u001b[32m2,073/2,120 \u001b[0m [ \u001b[33m0:00:04\u001b[0m < \u001b[36m0:00:01\u001b[0m , \u001b[31m773 it/s\u001b[0m ]t/s\u001b[0m ]\n",
            "\u001b[?25h\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m sat - Iteration \u001b[1;36m31\u001b[0m of \u001b[1;36m35\u001b[0m                                              \n",
            "\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m Accumulating statistics\u001b[33m...\u001b[0m                                            \n",
            "\u001b[2K\u001b[35m  98%\u001b[0m \u001b[91m━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m \u001b[32m2,071/2,120 \u001b[0m [ \u001b[33m0:00:05\u001b[0m < \u001b[36m0:00:01\u001b[0m , \u001b[31m627 it/s\u001b[0m ]t/s\u001b[0m ]\n",
            "\u001b[?25h\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m sat - Iteration \u001b[1;36m32\u001b[0m of \u001b[1;36m35\u001b[0m                                              \n",
            "\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m Accumulating statistics\u001b[33m...\u001b[0m                                            \n",
            "\u001b[2K\u001b[35m 100%\u001b[0m \u001b[91m━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[91m╸\u001b[0m \u001b[32m2,117/2,120 \u001b[0m [ \u001b[33m0:00:05\u001b[0m < \u001b[36m0:00:01\u001b[0m , \u001b[31m606 it/s\u001b[0m ]t/s\u001b[0m ]\n",
            "\u001b[?25h\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m sat - Iteration \u001b[1;36m33\u001b[0m of \u001b[1;36m35\u001b[0m                                              \n",
            "\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m Accumulating statistics\u001b[33m...\u001b[0m                                            \n",
            "\u001b[2K\u001b[35m 100%\u001b[0m \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m2,122/2,120 \u001b[0m [ \u001b[33m0:00:03\u001b[0m < \u001b[36m0:00:00\u001b[0m , \u001b[31m637 it/s\u001b[0m ]t/s\u001b[0m ]t/s\u001b[0m ]\n",
            "\u001b[?25h\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m sat - Iteration \u001b[1;36m34\u001b[0m of \u001b[1;36m35\u001b[0m                                              \n",
            "\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m Accumulating statistics\u001b[33m...\u001b[0m                                            \n",
            "\u001b[2K\u001b[35m 100%\u001b[0m \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m2,128/2,120 \u001b[0m [ \u001b[33m0:00:03\u001b[0m < \u001b[36m0:00:00\u001b[0m , \u001b[31m597 it/s\u001b[0m ]t/s\u001b[0m ]t/s\u001b[0m ]\n",
            "\u001b[?25h\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m sat - Iteration \u001b[1;36m35\u001b[0m of \u001b[1;36m35\u001b[0m                                              \n",
            "\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m Accumulating statistics\u001b[33m...\u001b[0m                                            \n",
            "\u001b[2K\u001b[35m 100%\u001b[0m \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m2,121/2,120 \u001b[0m [ \u001b[33m0:00:03\u001b[0m < \u001b[36m0:00:00\u001b[0m , \u001b[31m682 it/s\u001b[0m ]t/s\u001b[0m ]t/s\u001b[0m ]\n",
            "\u001b[?25h\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m Creating alignment model for speaker-independent features\u001b[33m...\u001b[0m          \n",
            "\u001b[2K\u001b[35m 100%\u001b[0m \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m2,127/2,120 \u001b[0m [ \u001b[33m0:00:05\u001b[0m < \u001b[36m0:00:00\u001b[0m , \u001b[31m405 it/s\u001b[0m ]t/s\u001b[0m ]t/s\u001b[0m ]\n",
            "\u001b[?25h\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m Training complete!                                                    \n",
            "\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m Compiling training graphs\u001b[33m...\u001b[0m                                          \n",
            "\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m Generating alignments\u001b[33m...\u001b[0m                                              \n",
            "\u001b[2K\u001b[35m  99%\u001b[0m \u001b[91m━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[91m╸\u001b[0m \u001b[32m2,106/2,120 \u001b[0m [ \u001b[33m0:00:16\u001b[0m < \u001b[36m0:00:01\u001b[0m , \u001b[31m145 it/s\u001b[0m ]t/s\u001b[0m ]\n",
            "\u001b[?25h\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m Calculating fMLLR for speaker adaptation\u001b[33m...\u001b[0m                           \n",
            "\u001b[2K\u001b[35m 100%\u001b[0m \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m11/1 \u001b[0m [ \u001b[33m0:00:07\u001b[0m < \u001b[36m0:00:00\u001b[0m , \u001b[31m11 it/s\u001b[0m ]\n",
            "\u001b[?25h\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m Generating alignments\u001b[33m...\u001b[0m                                              \n",
            "\u001b[2K\u001b[35m  99%\u001b[0m \u001b[91m━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[91m╸\u001b[0m \u001b[32m2,108/2,120 \u001b[0m [ \u001b[33m0:00:16\u001b[0m < \u001b[36m0:00:01\u001b[0m , \u001b[31m149 it/s\u001b[0m ]t/s\u001b[0m ]\n",
            "\u001b[?25h\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m Initializing training for sat_2\u001b[33m...\u001b[0m                                    \n",
            "\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m Accumulating tree stats\u001b[33m...\u001b[0m                                            \n",
            "\u001b[2K\u001b[35m  97%\u001b[0m \u001b[91m━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m \u001b[32m2,048/2,120 \u001b[0m [ \u001b[33m0:00:04\u001b[0m < \u001b[36m0:00:01\u001b[0m , \u001b[31m1,124 it/s\u001b[0m ]t/s\u001b[0m ]\n",
            "\u001b[?25h\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m Converting alignments\u001b[33m...\u001b[0m                                              \n",
            "\u001b[2K\u001b[35m 100%\u001b[0m \u001b[91m━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[91m╸\u001b[0m \u001b[32m2,116/2,120 \u001b[0m [ \u001b[33m0:00:02\u001b[0m < \u001b[36m0:00:01\u001b[0m , \u001b[31m14,610 it/s\u001b[0m ]t/s\u001b[0m ]\n",
            "\u001b[?25h\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m Compiling training graphs\u001b[33m...\u001b[0m                                          \n",
            "\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m Initialization complete!                                              \n",
            "\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m sat_2 - Iteration \u001b[1;36m1\u001b[0m of \u001b[1;36m35\u001b[0m                                             \n",
            "\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m Accumulating statistics\u001b[33m...\u001b[0m                                            \n",
            "\u001b[2K\u001b[35m  99%\u001b[0m \u001b[91m━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[91m╸\u001b[0m \u001b[32m2,100/2,120 \u001b[0m [ \u001b[33m0:00:04\u001b[0m < \u001b[36m0:00:01\u001b[0m , \u001b[31m774 it/s\u001b[0m ]t/s\u001b[0m ]\n",
            "\u001b[?25h\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m sat_2 - Iteration \u001b[1;36m2\u001b[0m of \u001b[1;36m35\u001b[0m                                             \n",
            "\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m Calculating fMLLR for speaker adaptation\u001b[33m...\u001b[0m                           \n",
            "\u001b[2K\u001b[35m 100%\u001b[0m \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m10/1 \u001b[0m [ \u001b[33m0:00:06\u001b[0m < \u001b[36m0:00:00\u001b[0m , \u001b[31m8 it/s\u001b[0m ]\n",
            "\u001b[?25h\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m Accumulating statistics\u001b[33m...\u001b[0m                                            \n",
            "\u001b[2K\u001b[35m  98%\u001b[0m \u001b[91m━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m \u001b[32m2,081/2,120 \u001b[0m [ \u001b[33m0:00:05\u001b[0m < \u001b[36m0:00:01\u001b[0m , \u001b[31m704 it/s\u001b[0m ]t/s\u001b[0m ]\n",
            "\u001b[?25h\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m sat_2 - Iteration \u001b[1;36m3\u001b[0m of \u001b[1;36m35\u001b[0m                                             \n",
            "\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m Accumulating statistics\u001b[33m...\u001b[0m                                            \n",
            "\u001b[2K\u001b[35m  99%\u001b[0m \u001b[91m━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[91m╸\u001b[0m \u001b[32m2,096/2,120 \u001b[0m [ \u001b[33m0:00:05\u001b[0m < \u001b[36m0:00:01\u001b[0m , \u001b[31m718 it/s\u001b[0m ]t/s\u001b[0m ]\n",
            "\u001b[?25h\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m sat_2 - Iteration \u001b[1;36m4\u001b[0m of \u001b[1;36m35\u001b[0m                                             \n",
            "\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m Calculating fMLLR for speaker adaptation\u001b[33m...\u001b[0m                           \n",
            "\u001b[2K\u001b[35m 100%\u001b[0m \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m11/1 \u001b[0m [ \u001b[33m0:00:06\u001b[0m < \u001b[36m0:00:00\u001b[0m , \u001b[31m10 it/s\u001b[0m ]\n",
            "\u001b[?25h\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m Accumulating statistics\u001b[33m...\u001b[0m                                            \n",
            "\u001b[2K\u001b[35m 100%\u001b[0m \u001b[91m━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[91m╸\u001b[0m \u001b[32m2,115/2,120 \u001b[0m [ \u001b[33m0:00:05\u001b[0m < \u001b[36m0:00:01\u001b[0m , \u001b[31m637 it/s\u001b[0m ]t/s\u001b[0m ]\n",
            "\u001b[?25h\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m sat_2 - Iteration \u001b[1;36m5\u001b[0m of \u001b[1;36m35\u001b[0m                                             \n",
            "\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m Accumulating statistics\u001b[33m...\u001b[0m                                            \n",
            "\u001b[2K\u001b[35m  98%\u001b[0m \u001b[91m━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m \u001b[32m2,076/2,120 \u001b[0m [ \u001b[33m0:00:05\u001b[0m < \u001b[36m0:00:01\u001b[0m , \u001b[31m686 it/s\u001b[0m ]t/s\u001b[0m ]\n",
            "\u001b[?25h\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m sat_2 - Iteration \u001b[1;36m6\u001b[0m of \u001b[1;36m35\u001b[0m                                             \n",
            "\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m Calculating fMLLR for speaker adaptation\u001b[33m...\u001b[0m                           \n",
            "\u001b[2K\u001b[35m 100%\u001b[0m \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m12/1 \u001b[0m [ \u001b[33m0:00:06\u001b[0m < \u001b[36m0:00:00\u001b[0m , \u001b[31m10 it/s\u001b[0m ]\n",
            "\u001b[?25h\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m Accumulating statistics\u001b[33m...\u001b[0m                                            \n",
            "\u001b[2K\u001b[35m  98%\u001b[0m \u001b[91m━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[91m╸\u001b[0m \u001b[32m2,085/2,120 \u001b[0m [ \u001b[33m0:00:05\u001b[0m < \u001b[36m0:00:01\u001b[0m , \u001b[31m566 it/s\u001b[0m ]t/s\u001b[0m ]\n",
            "\u001b[?25h\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m sat_2 - Iteration \u001b[1;36m7\u001b[0m of \u001b[1;36m35\u001b[0m                                             \n",
            "\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m Accumulating statistics\u001b[33m...\u001b[0m                                            \n",
            "\u001b[2K\u001b[35m 100%\u001b[0m \u001b[91m━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[91m╸\u001b[0m \u001b[32m2,114/2,120 \u001b[0m [ \u001b[33m0:00:05\u001b[0m < \u001b[36m0:00:01\u001b[0m , \u001b[31m595 it/s\u001b[0m ]t/s\u001b[0m ]\n",
            "\u001b[?25h\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m sat_2 - Iteration \u001b[1;36m8\u001b[0m of \u001b[1;36m35\u001b[0m                                             \n",
            "\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m Accumulating statistics\u001b[33m...\u001b[0m                                            \n",
            "\u001b[2K\u001b[35m  99%\u001b[0m \u001b[91m━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[91m╸\u001b[0m \u001b[32m2,092/2,120 \u001b[0m [ \u001b[33m0:00:05\u001b[0m < \u001b[36m0:00:01\u001b[0m , \u001b[31m652 it/s\u001b[0m ]t/s\u001b[0m ]\n",
            "\u001b[?25h\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m sat_2 - Iteration \u001b[1;36m9\u001b[0m of \u001b[1;36m35\u001b[0m                                             \n",
            "\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m Accumulating statistics\u001b[33m...\u001b[0m                                            \n",
            "\u001b[2K\u001b[35m 100%\u001b[0m \u001b[91m━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[91m╸\u001b[0m \u001b[32m2,117/2,120 \u001b[0m [ \u001b[33m0:00:05\u001b[0m < \u001b[36m0:00:01\u001b[0m , \u001b[31m674 it/s\u001b[0m ]t/s\u001b[0m ]\n",
            "\u001b[?25h\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m sat_2 - Iteration \u001b[1;36m10\u001b[0m of \u001b[1;36m35\u001b[0m                                            \n",
            "\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m Generating alignments\u001b[33m...\u001b[0m                                              \n",
            "\u001b[2K\u001b[35m  99%\u001b[0m \u001b[91m━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[91m╸\u001b[0m \u001b[32m2,095/2,120 \u001b[0m [ \u001b[33m0:00:15\u001b[0m < \u001b[36m0:00:01\u001b[0m , \u001b[31m162 it/s\u001b[0m ]t/s\u001b[0m ]\n",
            "\u001b[?25h\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m Accumulating statistics\u001b[33m...\u001b[0m                                            \n",
            "\u001b[2K\u001b[35m  98%\u001b[0m \u001b[91m━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m \u001b[32m2,070/2,120 \u001b[0m [ \u001b[33m0:00:05\u001b[0m < \u001b[36m0:00:01\u001b[0m , \u001b[31m624 it/s\u001b[0m ]t/s\u001b[0m ]\n",
            "\u001b[?25h\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m sat_2 - Iteration \u001b[1;36m11\u001b[0m of \u001b[1;36m35\u001b[0m                                            \n",
            "\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m Accumulating statistics\u001b[33m...\u001b[0m                                            \n",
            "\u001b[2K\u001b[35m  98%\u001b[0m \u001b[91m━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m \u001b[32m2,079/2,120 \u001b[0m [ \u001b[33m0:00:05\u001b[0m < \u001b[36m0:00:01\u001b[0m , \u001b[31m610 it/s\u001b[0m ]t/s\u001b[0m ]\n",
            "\u001b[?25h\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m sat_2 - Iteration \u001b[1;36m12\u001b[0m of \u001b[1;36m35\u001b[0m                                            \n",
            "\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m Calculating fMLLR for speaker adaptation\u001b[33m...\u001b[0m                           \n",
            "\u001b[2K\u001b[35m 100%\u001b[0m \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m12/1 \u001b[0m [ \u001b[33m0:00:07\u001b[0m < \u001b[36m0:00:00\u001b[0m , \u001b[31m10 it/s\u001b[0m ]\n",
            "\u001b[?25h\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m Accumulating statistics\u001b[33m...\u001b[0m                                            \n",
            "\u001b[2K\u001b[35m  99%\u001b[0m \u001b[91m━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[91m╸\u001b[0m \u001b[32m2,109/2,120 \u001b[0m [ \u001b[33m0:00:05\u001b[0m < \u001b[36m0:00:01\u001b[0m , \u001b[31m706 it/s\u001b[0m ]t/s\u001b[0m ]\n",
            "\u001b[?25h\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m sat_2 - Iteration \u001b[1;36m13\u001b[0m of \u001b[1;36m35\u001b[0m                                            \n",
            "\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m Accumulating statistics\u001b[33m...\u001b[0m                                            \n",
            "\u001b[2K\u001b[35m  98%\u001b[0m \u001b[91m━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[91m╸\u001b[0m \u001b[32m2,084/2,120 \u001b[0m [ \u001b[33m0:00:06\u001b[0m < \u001b[36m0:00:01\u001b[0m , \u001b[31m529 it/s\u001b[0m ]t/s\u001b[0m ]\n",
            "\u001b[?25h\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m sat_2 - Iteration \u001b[1;36m14\u001b[0m of \u001b[1;36m35\u001b[0m                                            \n",
            "\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m Accumulating statistics\u001b[33m...\u001b[0m                                            \n",
            "\u001b[2K\u001b[35m  99%\u001b[0m \u001b[91m━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[91m╸\u001b[0m \u001b[32m2,094/2,120 \u001b[0m [ \u001b[33m0:00:05\u001b[0m < \u001b[36m0:00:01\u001b[0m , \u001b[31m621 it/s\u001b[0m ]t/s\u001b[0m ]\n",
            "\u001b[?25h\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m sat_2 - Iteration \u001b[1;36m15\u001b[0m of \u001b[1;36m35\u001b[0m                                            \n",
            "\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m Accumulating statistics\u001b[33m...\u001b[0m                                            \n",
            "\u001b[2K\u001b[35m  98%\u001b[0m \u001b[91m━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m \u001b[32m2,078/2,120 \u001b[0m [ \u001b[33m0:00:05\u001b[0m < \u001b[36m0:00:01\u001b[0m , \u001b[31m559 it/s\u001b[0m ]t/s\u001b[0m ]\n",
            "\u001b[?25h\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m sat_2 - Iteration \u001b[1;36m16\u001b[0m of \u001b[1;36m35\u001b[0m                                            \n",
            "\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m Accumulating statistics\u001b[33m...\u001b[0m                                            \n",
            "\u001b[2K\u001b[35m 100%\u001b[0m \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m2,125/2,120 \u001b[0m [ \u001b[33m0:00:03\u001b[0m < \u001b[36m0:00:00\u001b[0m , \u001b[31m557 it/s\u001b[0m ]t/s\u001b[0m ]t/s\u001b[0m ]\n",
            "\u001b[?25h\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m sat_2 - Iteration \u001b[1;36m17\u001b[0m of \u001b[1;36m35\u001b[0m                                            \n",
            "\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m Accumulating statistics\u001b[33m...\u001b[0m                                            \n",
            "\u001b[2K\u001b[35m 100%\u001b[0m \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m2,120/2,120 \u001b[0m [ \u001b[33m0:00:03\u001b[0m < \u001b[36m0:00:00\u001b[0m , \u001b[31m653 it/s\u001b[0m ][31m662 it/s\u001b[0m ]\n",
            "\u001b[?25h\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m sat_2 - Iteration \u001b[1;36m18\u001b[0m of \u001b[1;36m35\u001b[0m                                            \n",
            "\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m Accumulating statistics\u001b[33m...\u001b[0m                                            \n",
            "\u001b[2K\u001b[35m 100%\u001b[0m \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m2,123/2,120 \u001b[0m [ \u001b[33m0:00:04\u001b[0m < \u001b[36m0:00:00\u001b[0m , \u001b[31m511 it/s\u001b[0m ]t/s\u001b[0m ]t/s\u001b[0m ]\n",
            "\u001b[?25h\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m sat_2 - Iteration \u001b[1;36m19\u001b[0m of \u001b[1;36m35\u001b[0m                                            \n",
            "\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m Accumulating statistics\u001b[33m...\u001b[0m                                            \n",
            "\u001b[2K\u001b[35m  99%\u001b[0m \u001b[91m━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[91m╸\u001b[0m \u001b[32m2,093/2,120 \u001b[0m [ \u001b[33m0:00:06\u001b[0m < \u001b[36m0:00:01\u001b[0m , \u001b[31m515 it/s\u001b[0m ]t/s\u001b[0m ]\n",
            "\u001b[?25h\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m sat_2 - Iteration \u001b[1;36m20\u001b[0m of \u001b[1;36m35\u001b[0m                                            \n",
            "\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m Generating alignments\u001b[33m...\u001b[0m                                              \n",
            "\u001b[2K\u001b[35m 100%\u001b[0m \u001b[91m━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[91m╸\u001b[0m \u001b[32m2,113/2,120 \u001b[0m [ \u001b[33m0:00:17\u001b[0m < \u001b[36m0:00:01\u001b[0m , \u001b[31m136 it/s\u001b[0m ]t/s\u001b[0m ]\n",
            "\u001b[?25h\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m Accumulating statistics\u001b[33m...\u001b[0m                                            \n",
            "\u001b[2K\u001b[35m  98%\u001b[0m \u001b[91m━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[91m╸\u001b[0m \u001b[32m2,085/2,120 \u001b[0m [ \u001b[33m0:00:06\u001b[0m < \u001b[36m0:00:01\u001b[0m , \u001b[31m528 it/s\u001b[0m ]t/s\u001b[0m ]\n",
            "\u001b[?25h\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m sat_2 - Iteration \u001b[1;36m21\u001b[0m of \u001b[1;36m35\u001b[0m                                            \n",
            "\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m Accumulating statistics\u001b[33m...\u001b[0m                                            \n",
            "\u001b[2K\u001b[35m 100%\u001b[0m \u001b[91m━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[91m╸\u001b[0m \u001b[32m2,115/2,120 \u001b[0m [ \u001b[33m0:00:06\u001b[0m < \u001b[36m0:00:01\u001b[0m , \u001b[31m508 it/s\u001b[0m ]t/s\u001b[0m ]\n",
            "\u001b[?25h\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m sat_2 - Iteration \u001b[1;36m22\u001b[0m of \u001b[1;36m35\u001b[0m                                            \n",
            "\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m Accumulating statistics\u001b[33m...\u001b[0m                                            \n",
            "\u001b[2K\u001b[35m 100%\u001b[0m \u001b[91m━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[91m╸\u001b[0m \u001b[32m2,112/2,120 \u001b[0m [ \u001b[33m0:00:06\u001b[0m < \u001b[36m0:00:01\u001b[0m , \u001b[31m475 it/s\u001b[0m ]t/s\u001b[0m ]\n",
            "\u001b[?25h\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m sat_2 - Iteration \u001b[1;36m23\u001b[0m of \u001b[1;36m35\u001b[0m                                            \n",
            "\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m Accumulating statistics\u001b[33m...\u001b[0m                                            \n",
            "\u001b[2K\u001b[35m  99%\u001b[0m \u001b[91m━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[91m╸\u001b[0m \u001b[32m2,102/2,120 \u001b[0m [ \u001b[33m0:00:06\u001b[0m < \u001b[36m0:00:01\u001b[0m , \u001b[31m456 it/s\u001b[0m ]t/s\u001b[0m ]\n",
            "\u001b[?25h\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m sat_2 - Iteration \u001b[1;36m24\u001b[0m of \u001b[1;36m35\u001b[0m                                            \n",
            "\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m Accumulating statistics\u001b[33m...\u001b[0m                                            \n",
            "\u001b[2K\u001b[35m  99%\u001b[0m \u001b[91m━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[91m╸\u001b[0m \u001b[32m2,090/2,120 \u001b[0m [ \u001b[33m0:00:07\u001b[0m < \u001b[36m0:00:01\u001b[0m , \u001b[31m423 it/s\u001b[0m ]t/s\u001b[0m ]\n",
            "\u001b[?25h\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m sat_2 - Iteration \u001b[1;36m25\u001b[0m of \u001b[1;36m35\u001b[0m                                            \n",
            "\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m Accumulating statistics\u001b[33m...\u001b[0m                                            \n",
            "\u001b[2K\u001b[35m 100%\u001b[0m \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m2,124/2,120 \u001b[0m [ \u001b[33m0:00:05\u001b[0m < \u001b[36m0:00:00\u001b[0m , \u001b[31m434 it/s\u001b[0m ]t/s\u001b[0m ]t/s\u001b[0m ]\n",
            "\u001b[?25h\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m sat_2 - Iteration \u001b[1;36m26\u001b[0m of \u001b[1;36m35\u001b[0m                                            \n",
            "\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m Accumulating statistics\u001b[33m...\u001b[0m                                            \n",
            "\u001b[2K\u001b[35m 100%\u001b[0m \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m2,123/2,120 \u001b[0m [ \u001b[33m0:00:05\u001b[0m < \u001b[36m0:00:00\u001b[0m , \u001b[31m403 it/s\u001b[0m ]t/s\u001b[0m ]t/s\u001b[0m ]\n",
            "\u001b[?25h\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m sat_2 - Iteration \u001b[1;36m27\u001b[0m of \u001b[1;36m35\u001b[0m                                            \n",
            "\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m Accumulating statistics\u001b[33m...\u001b[0m                                            \n",
            "\u001b[2K\u001b[35m  99%\u001b[0m \u001b[91m━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[91m╸\u001b[0m \u001b[32m2,098/2,120 \u001b[0m [ \u001b[33m0:00:08\u001b[0m < \u001b[36m0:00:01\u001b[0m , \u001b[31m360 it/s\u001b[0m ]t/s\u001b[0m ]\n",
            "\u001b[?25h\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m sat_2 - Iteration \u001b[1;36m28\u001b[0m of \u001b[1;36m35\u001b[0m                                            \n",
            "\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m Accumulating statistics\u001b[33m...\u001b[0m                                            \n",
            "\u001b[2K\u001b[35m  99%\u001b[0m \u001b[91m━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[91m╸\u001b[0m \u001b[32m2,103/2,120 \u001b[0m [ \u001b[33m0:00:07\u001b[0m < \u001b[36m0:00:01\u001b[0m , \u001b[31m401 it/s\u001b[0m ]t/s\u001b[0m ]\n",
            "\u001b[?25h\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m sat_2 - Iteration \u001b[1;36m29\u001b[0m of \u001b[1;36m35\u001b[0m                                            \n",
            "\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m Accumulating statistics\u001b[33m...\u001b[0m                                            \n",
            "\u001b[2K\u001b[35m  99%\u001b[0m \u001b[91m━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[91m╸\u001b[0m \u001b[32m2,097/2,120 \u001b[0m [ \u001b[33m0:00:08\u001b[0m < \u001b[36m0:00:01\u001b[0m , \u001b[31m347 it/s\u001b[0m ]t/s\u001b[0m ]\n",
            "\u001b[?25h\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m sat_2 - Iteration \u001b[1;36m30\u001b[0m of \u001b[1;36m35\u001b[0m                                            \n",
            "\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m Generating alignments\u001b[33m...\u001b[0m                                              \n",
            "\u001b[2K\u001b[35m 100%\u001b[0m \u001b[91m━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[91m╸\u001b[0m \u001b[32m2,112/2,120 \u001b[0m [ \u001b[33m0:00:18\u001b[0m < \u001b[36m0:00:01\u001b[0m , \u001b[31m128 it/s\u001b[0m ]t/s\u001b[0m ]\n",
            "\u001b[?25h\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m Accumulating statistics\u001b[33m...\u001b[0m                                            \n",
            "\u001b[2K\u001b[35m 100%\u001b[0m \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m2,128/2,120 \u001b[0m [ \u001b[33m0:00:05\u001b[0m < \u001b[36m0:00:00\u001b[0m , \u001b[31m365 it/s\u001b[0m ]t/s\u001b[0m ]t/s\u001b[0m ]\n",
            "\u001b[?25h\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m sat_2 - Iteration \u001b[1;36m31\u001b[0m of \u001b[1;36m35\u001b[0m                                            \n",
            "\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m Accumulating statistics\u001b[33m...\u001b[0m                                            \n",
            "\u001b[2K\u001b[35m  99%\u001b[0m \u001b[91m━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[91m╸\u001b[0m \u001b[32m2,094/2,120 \u001b[0m [ \u001b[33m0:00:08\u001b[0m < \u001b[36m0:00:01\u001b[0m , \u001b[31m354 it/s\u001b[0m ]t/s\u001b[0m ]\n",
            "\u001b[?25h\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m sat_2 - Iteration \u001b[1;36m32\u001b[0m of \u001b[1;36m35\u001b[0m                                            \n",
            "\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m Accumulating statistics\u001b[33m...\u001b[0m                                            \n",
            "\u001b[2K\u001b[35m  99%\u001b[0m \u001b[91m━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[91m╸\u001b[0m \u001b[32m2,109/2,120 \u001b[0m [ \u001b[33m0:00:08\u001b[0m < \u001b[36m0:00:01\u001b[0m , \u001b[31m350 it/s\u001b[0m ]t/s\u001b[0m ]\n",
            "\u001b[?25h\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m sat_2 - Iteration \u001b[1;36m33\u001b[0m of \u001b[1;36m35\u001b[0m                                            \n",
            "\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m Accumulating statistics\u001b[33m...\u001b[0m                                            \n",
            "\u001b[2K\u001b[35m  99%\u001b[0m \u001b[91m━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[91m╸\u001b[0m \u001b[32m2,103/2,120 \u001b[0m [ \u001b[33m0:00:08\u001b[0m < \u001b[36m0:00:01\u001b[0m , \u001b[31m345 it/s\u001b[0m ]t/s\u001b[0m ]\n",
            "\u001b[?25h\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m sat_2 - Iteration \u001b[1;36m34\u001b[0m of \u001b[1;36m35\u001b[0m                                            \n",
            "\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m Accumulating statistics\u001b[33m...\u001b[0m                                            \n",
            "\u001b[2K\u001b[35m  99%\u001b[0m \u001b[91m━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[91m╸\u001b[0m \u001b[32m2,103/2,120 \u001b[0m [ \u001b[33m0:00:08\u001b[0m < \u001b[36m0:00:01\u001b[0m , \u001b[31m344 it/s\u001b[0m ]t/s\u001b[0m ]\n",
            "\u001b[?25h\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m sat_2 - Iteration \u001b[1;36m35\u001b[0m of \u001b[1;36m35\u001b[0m                                            \n",
            "\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m Accumulating statistics\u001b[33m...\u001b[0m                                            \n",
            "\u001b[2K\u001b[35m  99%\u001b[0m \u001b[91m━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[91m╸\u001b[0m \u001b[32m2,093/2,120 \u001b[0m [ \u001b[33m0:00:07\u001b[0m < \u001b[36m0:00:01\u001b[0m , \u001b[31m371 it/s\u001b[0m ]t/s\u001b[0m ]\n",
            "\u001b[?25h\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m Creating alignment model for speaker-independent features\u001b[33m...\u001b[0m          \n",
            "\u001b[2K\u001b[35m  99%\u001b[0m \u001b[91m━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[91m╸\u001b[0m \u001b[32m2,109/2,120 \u001b[0m [ \u001b[33m0:00:09\u001b[0m < \u001b[36m0:00:01\u001b[0m , \u001b[31m299 it/s\u001b[0m ]t/s\u001b[0m ]\n",
            "\u001b[?25h\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m Training complete!                                                    \n",
            "\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m Compiling training graphs\u001b[33m...\u001b[0m                                          \n",
            "\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m Generating alignments\u001b[33m...\u001b[0m                                              \n",
            "\u001b[2K\u001b[35m  99%\u001b[0m \u001b[91m━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[91m╸\u001b[0m \u001b[32m2,106/2,120 \u001b[0m [ \u001b[33m0:00:18\u001b[0m < \u001b[36m0:00:01\u001b[0m , \u001b[31m127 it/s\u001b[0m ]t/s\u001b[0m ]\n",
            "\u001b[?25h\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m Calculating fMLLR for speaker adaptation\u001b[33m...\u001b[0m                           \n",
            "\u001b[2K\u001b[35m 100%\u001b[0m \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m11/1 \u001b[0m [ \u001b[33m0:00:07\u001b[0m < \u001b[36m0:00:00\u001b[0m , \u001b[31m8 it/s\u001b[0m ]\n",
            "\u001b[?25h\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m Generating alignments\u001b[33m...\u001b[0m                                              \n",
            "\u001b[2K\u001b[35m 100%\u001b[0m \u001b[91m━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[91m╸\u001b[0m \u001b[32m2,111/2,120 \u001b[0m [ \u001b[33m0:00:17\u001b[0m < \u001b[36m0:00:01\u001b[0m , \u001b[31m134 it/s\u001b[0m ]t/s\u001b[0m ]\n",
            "\u001b[?25h\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m Generating pronunciations\u001b[33m...\u001b[0m                                          \n",
            "\u001b[2K\u001b[35m  98%\u001b[0m \u001b[91m━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m \u001b[32m2,073/2,120 \u001b[0m [ \u001b[33m0:00:04\u001b[0m < \u001b[36m0:00:01\u001b[0m , \u001b[31m1,031 it/s\u001b[0m ]t/s\u001b[0m ]\n",
            "\u001b[?25h\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m Compiling training graphs\u001b[33m...\u001b[0m                                          \n",
            "\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m Generating alignments\u001b[33m...\u001b[0m                                              \n",
            "\u001b[2K\u001b[35m 100%\u001b[0m \u001b[91m━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[91m╸\u001b[0m \u001b[32m2,116/2,120 \u001b[0m [ \u001b[33m0:00:18\u001b[0m < \u001b[36m0:00:01\u001b[0m , \u001b[31m131 it/s\u001b[0m ]t/s\u001b[0m ]\n",
            "\u001b[?25h\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m Calculating fMLLR for speaker adaptation\u001b[33m...\u001b[0m                           \n",
            "\u001b[2K\u001b[35m 100%\u001b[0m \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m10/1 \u001b[0m [ \u001b[33m0:00:07\u001b[0m < \u001b[36m0:00:00\u001b[0m , \u001b[31m11 it/s\u001b[0m ]\n",
            "\u001b[?25h\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m Generating alignments\u001b[33m...\u001b[0m                                              \n",
            "\u001b[2K\u001b[35m 100%\u001b[0m \u001b[91m━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[91m╸\u001b[0m \u001b[32m2,116/2,120 \u001b[0m [ \u001b[33m0:00:18\u001b[0m < \u001b[36m0:00:01\u001b[0m , \u001b[31m127 it/s\u001b[0m ]t/s\u001b[0m ]\n",
            "\u001b[?25h\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m Initializing training for sat_3\u001b[33m...\u001b[0m                                    \n",
            "\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m Accumulating tree stats\u001b[33m...\u001b[0m                                            \n",
            "\u001b[2K\u001b[35m  97%\u001b[0m \u001b[91m━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m \u001b[32m2,060/2,120 \u001b[0m [ \u001b[33m0:00:04\u001b[0m < \u001b[36m0:00:01\u001b[0m , \u001b[31m1,103 it/s\u001b[0m ]t/s\u001b[0m ]\n",
            "\u001b[?25h\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m Converting alignments\u001b[33m...\u001b[0m                                              \n",
            "\u001b[2K\u001b[35m  94%\u001b[0m \u001b[91m━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[91m╸\u001b[0m\u001b[90m━\u001b[0m \u001b[32m2,003/2,120 \u001b[0m [ \u001b[33m0:00:02\u001b[0m < \u001b[36m0:00:01\u001b[0m , \u001b[31m15,739 it/s\u001b[0m ]\n",
            "\u001b[?25h\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m Compiling training graphs\u001b[33m...\u001b[0m                                          \n",
            "\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m Initialization complete!                                              \n",
            "\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m sat_3 - Iteration \u001b[1;36m1\u001b[0m of \u001b[1;36m35\u001b[0m                                             \n",
            "\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m Accumulating statistics\u001b[33m...\u001b[0m                                            \n",
            "\u001b[2K\u001b[35m  98%\u001b[0m \u001b[91m━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m \u001b[32m2,073/2,120 \u001b[0m [ \u001b[33m0:00:04\u001b[0m < \u001b[36m0:00:01\u001b[0m , \u001b[31m754 it/s\u001b[0m ]t/s\u001b[0m ]\n",
            "\u001b[?25h\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m sat_3 - Iteration \u001b[1;36m2\u001b[0m of \u001b[1;36m35\u001b[0m                                             \n",
            "\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m Calculating fMLLR for speaker adaptation\u001b[33m...\u001b[0m                           \n",
            "\u001b[2K\u001b[35m 100%\u001b[0m \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m10/1 \u001b[0m [ \u001b[33m0:00:06\u001b[0m < \u001b[36m0:00:00\u001b[0m , \u001b[31m7 it/s\u001b[0m ]\n",
            "\u001b[?25h\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m Accumulating statistics\u001b[33m...\u001b[0m                                            \n",
            "\u001b[2K\u001b[35m 100%\u001b[0m \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m2,120/2,120 \u001b[0m [ \u001b[33m0:00:02\u001b[0m < \u001b[36m0:00:00\u001b[0m , \u001b[31m771 it/s\u001b[0m ][31m751 it/s\u001b[0m ]\n",
            "\u001b[?25h\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m sat_3 - Iteration \u001b[1;36m3\u001b[0m of \u001b[1;36m35\u001b[0m                                             \n",
            "\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m Accumulating statistics\u001b[33m...\u001b[0m                                            \n",
            "\u001b[2K\u001b[35m  98%\u001b[0m \u001b[91m━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m \u001b[32m2,081/2,120 \u001b[0m [ \u001b[33m0:00:05\u001b[0m < \u001b[36m0:00:01\u001b[0m , \u001b[31m689 it/s\u001b[0m ]t/s\u001b[0m ]\n",
            "\u001b[?25h\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m sat_3 - Iteration \u001b[1;36m4\u001b[0m of \u001b[1;36m35\u001b[0m                                             \n",
            "\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m Calculating fMLLR for speaker adaptation\u001b[33m...\u001b[0m                           \n",
            "\u001b[2K\u001b[35m 100%\u001b[0m \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m10/1 \u001b[0m [ \u001b[33m0:00:05\u001b[0m < \u001b[36m0:00:00\u001b[0m , \u001b[31m8 it/s\u001b[0m ]\n",
            "\u001b[?25h\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m Accumulating statistics\u001b[33m...\u001b[0m                                            \n",
            "\u001b[2K\u001b[35m  99%\u001b[0m \u001b[91m━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[91m╸\u001b[0m \u001b[32m2,099/2,120 \u001b[0m [ \u001b[33m0:00:05\u001b[0m < \u001b[36m0:00:01\u001b[0m , \u001b[31m652 it/s\u001b[0m ]t/s\u001b[0m ]\n",
            "\u001b[?25h\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m sat_3 - Iteration \u001b[1;36m5\u001b[0m of \u001b[1;36m35\u001b[0m                                             \n",
            "\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m Accumulating statistics\u001b[33m...\u001b[0m                                            \n",
            "\u001b[2K\u001b[35m 100%\u001b[0m \u001b[91m━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[91m╸\u001b[0m \u001b[32m2,118/2,120 \u001b[0m [ \u001b[33m0:00:05\u001b[0m < \u001b[36m0:00:01\u001b[0m , \u001b[31m648 it/s\u001b[0m ]t/s\u001b[0m ]\n",
            "\u001b[?25h\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m sat_3 - Iteration \u001b[1;36m6\u001b[0m of \u001b[1;36m35\u001b[0m                                             \n",
            "\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m Calculating fMLLR for speaker adaptation\u001b[33m...\u001b[0m                           \n",
            "\u001b[2K\u001b[35m 100%\u001b[0m \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m12/1 \u001b[0m [ \u001b[33m0:00:07\u001b[0m < \u001b[36m0:00:00\u001b[0m , \u001b[31m9 it/s\u001b[0m ]\n",
            "\u001b[?25h\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m Accumulating statistics\u001b[33m...\u001b[0m                                            \n",
            "\u001b[2K\u001b[35m  98%\u001b[0m \u001b[91m━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m \u001b[32m2,070/2,120 \u001b[0m [ \u001b[33m0:00:05\u001b[0m < \u001b[36m0:00:01\u001b[0m , \u001b[31m559 it/s\u001b[0m ]t/s\u001b[0m ]\n",
            "\u001b[?25h\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m sat_3 - Iteration \u001b[1;36m7\u001b[0m of \u001b[1;36m35\u001b[0m                                             \n",
            "\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m Accumulating statistics\u001b[33m...\u001b[0m                                            \n",
            "\u001b[2K\u001b[35m  99%\u001b[0m \u001b[91m━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[91m╸\u001b[0m \u001b[32m2,103/2,120 \u001b[0m [ \u001b[33m0:00:06\u001b[0m < \u001b[36m0:00:01\u001b[0m , \u001b[31m535 it/s\u001b[0m ]t/s\u001b[0m ]\n",
            "\u001b[?25h\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m sat_3 - Iteration \u001b[1;36m8\u001b[0m of \u001b[1;36m35\u001b[0m                                             \n",
            "\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m Accumulating statistics\u001b[33m...\u001b[0m                                            \n",
            "\u001b[2K\u001b[35m  99%\u001b[0m \u001b[91m━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[91m╸\u001b[0m \u001b[32m2,108/2,120 \u001b[0m [ \u001b[33m0:00:06\u001b[0m < \u001b[36m0:00:01\u001b[0m , \u001b[31m504 it/s\u001b[0m ]t/s\u001b[0m ]\n",
            "\u001b[?25h\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m sat_3 - Iteration \u001b[1;36m9\u001b[0m of \u001b[1;36m35\u001b[0m                                             \n",
            "\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m Accumulating statistics\u001b[33m...\u001b[0m                                            \n",
            "\u001b[2K\u001b[35m 100%\u001b[0m \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m2,125/2,120 \u001b[0m [ \u001b[33m0:00:04\u001b[0m < \u001b[36m0:00:00\u001b[0m , \u001b[31m486 it/s\u001b[0m ]t/s\u001b[0m ]t/s\u001b[0m ]\n",
            "\u001b[?25h\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m sat_3 - Iteration \u001b[1;36m10\u001b[0m of \u001b[1;36m35\u001b[0m                                            \n",
            "\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m Generating alignments\u001b[33m...\u001b[0m                                              \n",
            "\u001b[2K\u001b[35m 100%\u001b[0m \u001b[91m━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[91m╸\u001b[0m \u001b[32m2,114/2,120 \u001b[0m [ \u001b[33m0:00:17\u001b[0m < \u001b[36m0:00:01\u001b[0m , \u001b[31m137 it/s\u001b[0m ]t/s\u001b[0m ]\n",
            "\u001b[?25h\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m Accumulating statistics\u001b[33m...\u001b[0m                                            \n",
            "\u001b[2K\u001b[35m 100%\u001b[0m \u001b[91m━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[91m╸\u001b[0m \u001b[32m2,110/2,120 \u001b[0m [ \u001b[33m0:00:07\u001b[0m < \u001b[36m0:00:01\u001b[0m , \u001b[31m418 it/s\u001b[0m ]t/s\u001b[0m ]\n",
            "\u001b[?25h\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m sat_3 - Iteration \u001b[1;36m11\u001b[0m of \u001b[1;36m35\u001b[0m                                            \n",
            "\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m Accumulating statistics\u001b[33m...\u001b[0m                                            \n",
            "\u001b[2K\u001b[35m  99%\u001b[0m \u001b[91m━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[91m╸\u001b[0m \u001b[32m2,101/2,120 \u001b[0m [ \u001b[33m0:00:07\u001b[0m < \u001b[36m0:00:01\u001b[0m , \u001b[31m377 it/s\u001b[0m ]t/s\u001b[0m ]\n",
            "\u001b[?25h\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m sat_3 - Iteration \u001b[1;36m12\u001b[0m of \u001b[1;36m35\u001b[0m                                            \n",
            "\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m Calculating fMLLR for speaker adaptation\u001b[33m...\u001b[0m                           \n",
            "\u001b[2K\u001b[35m 100%\u001b[0m \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m10/1 \u001b[0m [ \u001b[33m0:00:06\u001b[0m < \u001b[36m0:00:00\u001b[0m , \u001b[31m19 it/s\u001b[0m ]\n",
            "\u001b[?25h\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m Accumulating statistics\u001b[33m...\u001b[0m                                            \n",
            "\u001b[2K\u001b[35m 100%\u001b[0m \u001b[91m━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[91m╸\u001b[0m \u001b[32m2,114/2,120 \u001b[0m [ \u001b[33m0:00:06\u001b[0m < \u001b[36m0:00:01\u001b[0m , \u001b[31m468 it/s\u001b[0m ]t/s\u001b[0m ]\n",
            "\u001b[?25h\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m sat_3 - Iteration \u001b[1;36m13\u001b[0m of \u001b[1;36m35\u001b[0m                                            \n",
            "\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m Accumulating statistics\u001b[33m...\u001b[0m                                            \n",
            "\u001b[2K\u001b[35m 100%\u001b[0m \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m2,125/2,120 \u001b[0m [ \u001b[33m0:00:07\u001b[0m < \u001b[36m0:00:00\u001b[0m , \u001b[31m290 it/s\u001b[0m ]t/s\u001b[0m ]t/s\u001b[0m ]\n",
            "\u001b[?25h\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m sat_3 - Iteration \u001b[1;36m14\u001b[0m of \u001b[1;36m35\u001b[0m                                            \n",
            "\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m Accumulating statistics\u001b[33m...\u001b[0m                                            \n",
            "\u001b[2K\u001b[35m 100%\u001b[0m \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m2,124/2,120 \u001b[0m [ \u001b[33m0:00:15\u001b[0m < \u001b[36m0:00:00\u001b[0m , \u001b[31m137 it/s\u001b[0m ]t/s\u001b[0m ]t/s\u001b[0m ]\n",
            "\u001b[?25h\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m sat_3 - Iteration \u001b[1;36m15\u001b[0m of \u001b[1;36m35\u001b[0m                                            \n",
            "\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m Accumulating statistics\u001b[33m...\u001b[0m                                            \n",
            "\u001b[2K\u001b[35m 100%\u001b[0m \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m2,123/2,120 \u001b[0m [ \u001b[33m0:00:33\u001b[0m < \u001b[36m0:00:00\u001b[0m , \u001b[31m64 it/s\u001b[0m ]t/s\u001b[0m ]t/s\u001b[0m ]\n",
            "\u001b[?25h\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m sat_3 - Iteration \u001b[1;36m16\u001b[0m of \u001b[1;36m35\u001b[0m                                            \n",
            "\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m Accumulating statistics\u001b[33m...\u001b[0m                                            \n",
            "\u001b[2K\u001b[35m 100%\u001b[0m \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m2,128/2,120 \u001b[0m [ \u001b[33m0:00:38\u001b[0m < \u001b[36m0:00:00\u001b[0m , \u001b[31m55 it/s\u001b[0m ]t/s\u001b[0m ]t/s\u001b[0m ]\n",
            "\u001b[?25h\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m sat_3 - Iteration \u001b[1;36m17\u001b[0m of \u001b[1;36m35\u001b[0m                                            \n",
            "\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m Accumulating statistics\u001b[33m...\u001b[0m                                            \n",
            "\u001b[2K\u001b[35m 100%\u001b[0m \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m2,125/2,120 \u001b[0m [ \u001b[33m0:00:37\u001b[0m < \u001b[36m0:00:00\u001b[0m , \u001b[31m56 it/s\u001b[0m ]t/s\u001b[0m ]t/s\u001b[0m ]\n",
            "\u001b[?25h\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m sat_3 - Iteration \u001b[1;36m18\u001b[0m of \u001b[1;36m35\u001b[0m                                            \n",
            "\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m Accumulating statistics\u001b[33m...\u001b[0m                                            \n",
            "\u001b[2K\u001b[35m 100%\u001b[0m \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m2,126/2,120 \u001b[0m [ \u001b[33m0:00:38\u001b[0m < \u001b[36m0:00:00\u001b[0m , \u001b[31m56 it/s\u001b[0m ]t/s\u001b[0m ]t/s\u001b[0m ]\n",
            "\u001b[?25h\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m sat_3 - Iteration \u001b[1;36m19\u001b[0m of \u001b[1;36m35\u001b[0m                                            \n",
            "\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m Accumulating statistics\u001b[33m...\u001b[0m                                            \n",
            "\u001b[2K\u001b[35m 100%\u001b[0m \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m2,125/2,120 \u001b[0m [ \u001b[33m0:00:37\u001b[0m < \u001b[36m0:00:00\u001b[0m , \u001b[31m57 it/s\u001b[0m ]t/s\u001b[0m ]t/s\u001b[0m ]\n",
            "\u001b[?25h\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m sat_3 - Iteration \u001b[1;36m20\u001b[0m of \u001b[1;36m35\u001b[0m                                            \n",
            "\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m Generating alignments\u001b[33m...\u001b[0m                                              \n",
            "\u001b[2K\u001b[35m 100%\u001b[0m \u001b[91m━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[91m╸\u001b[0m \u001b[32m2,114/2,120 \u001b[0m [ \u001b[33m0:01:06\u001b[0m < \u001b[36m0:00:01\u001b[0m , \u001b[31m33 it/s\u001b[0m ]t/s\u001b[0m ]\n",
            "\u001b[?25h\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m Accumulating statistics\u001b[33m...\u001b[0m                                            \n",
            "\u001b[2K\u001b[35m 100%\u001b[0m \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m2,120/2,120 \u001b[0m [ \u001b[33m0:00:37\u001b[0m < \u001b[36m0:00:00\u001b[0m , \u001b[31m56 it/s\u001b[0m ]t/s\u001b[0m ]t/s\u001b[0m ]\n",
            "\u001b[?25h\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m sat_3 - Iteration \u001b[1;36m21\u001b[0m of \u001b[1;36m35\u001b[0m                                            \n",
            "\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m Accumulating statistics\u001b[33m...\u001b[0m                                            \n",
            "\u001b[2K\u001b[35m 100%\u001b[0m \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m2,122/2,120 \u001b[0m [ \u001b[33m0:00:38\u001b[0m < \u001b[36m0:00:00\u001b[0m , \u001b[31m56 it/s\u001b[0m ]t/s\u001b[0m ]t/s\u001b[0m ]\n",
            "\u001b[?25h\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m sat_3 - Iteration \u001b[1;36m22\u001b[0m of \u001b[1;36m35\u001b[0m                                            \n",
            "\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m Accumulating statistics\u001b[33m...\u001b[0m                                            \n",
            "\u001b[2K\u001b[35m 100%\u001b[0m \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m2,121/2,120 \u001b[0m [ \u001b[33m0:00:38\u001b[0m < \u001b[36m0:00:00\u001b[0m , \u001b[31m55 it/s\u001b[0m ]t/s\u001b[0m ]t/s\u001b[0m ]\n",
            "\u001b[?25h\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m sat_3 - Iteration \u001b[1;36m23\u001b[0m of \u001b[1;36m35\u001b[0m                                            \n",
            "\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m Accumulating statistics\u001b[33m...\u001b[0m                                            \n",
            "\u001b[2K\u001b[35m 100%\u001b[0m \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m2,127/2,120 \u001b[0m [ \u001b[33m0:00:38\u001b[0m < \u001b[36m0:00:00\u001b[0m , \u001b[31m55 it/s\u001b[0m ]t/s\u001b[0m ]t/s\u001b[0m ]\n",
            "\u001b[?25h\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m sat_3 - Iteration \u001b[1;36m24\u001b[0m of \u001b[1;36m35\u001b[0m                                            \n",
            "\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m Accumulating statistics\u001b[33m...\u001b[0m                                            \n",
            "\u001b[2K\u001b[35m 100%\u001b[0m \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m2,127/2,120 \u001b[0m [ \u001b[33m0:00:39\u001b[0m < \u001b[36m0:00:00\u001b[0m , \u001b[31m55 it/s\u001b[0m ]t/s\u001b[0m ]t/s\u001b[0m ]\n",
            "\u001b[?25h\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m sat_3 - Iteration \u001b[1;36m25\u001b[0m of \u001b[1;36m35\u001b[0m                                            \n",
            "\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m Accumulating statistics\u001b[33m...\u001b[0m                                            \n",
            "\u001b[2K\u001b[35m 100%\u001b[0m \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m2,123/2,120 \u001b[0m [ \u001b[33m0:00:39\u001b[0m < \u001b[36m0:00:00\u001b[0m , \u001b[31m54 it/s\u001b[0m ]t/s\u001b[0m ]t/s\u001b[0m ]\n",
            "\u001b[?25h\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m sat_3 - Iteration \u001b[1;36m26\u001b[0m of \u001b[1;36m35\u001b[0m                                            \n",
            "\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m Accumulating statistics\u001b[33m...\u001b[0m                                            \n",
            "\u001b[2K\u001b[35m 100%\u001b[0m \u001b[91m━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[91m╸\u001b[0m \u001b[32m2,119/2,120 \u001b[0m [ \u001b[33m0:00:41\u001b[0m < \u001b[36m0:00:01\u001b[0m , \u001b[31m55 it/s\u001b[0m ]t/s\u001b[0m ]\n",
            "\u001b[?25h\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m sat_3 - Iteration \u001b[1;36m27\u001b[0m of \u001b[1;36m35\u001b[0m                                            \n",
            "\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m Accumulating statistics\u001b[33m...\u001b[0m                                            \n",
            "\u001b[2K\u001b[35m 100%\u001b[0m \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m2,126/2,120 \u001b[0m [ \u001b[33m0:00:38\u001b[0m < \u001b[36m0:00:00\u001b[0m , \u001b[31m56 it/s\u001b[0m ]t/s\u001b[0m ]t/s\u001b[0m ]\n",
            "\u001b[?25h\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m sat_3 - Iteration \u001b[1;36m28\u001b[0m of \u001b[1;36m35\u001b[0m                                            \n",
            "\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m Accumulating statistics\u001b[33m...\u001b[0m                                            \n",
            "\u001b[2K\u001b[35m 100%\u001b[0m \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m2,126/2,120 \u001b[0m [ \u001b[33m0:00:39\u001b[0m < \u001b[36m0:00:00\u001b[0m , \u001b[31m55 it/s\u001b[0m ]t/s\u001b[0m ]t/s\u001b[0m ]\n",
            "\u001b[?25h\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m sat_3 - Iteration \u001b[1;36m29\u001b[0m of \u001b[1;36m35\u001b[0m                                            \n",
            "\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m Accumulating statistics\u001b[33m...\u001b[0m                                            \n",
            "\u001b[2K\u001b[35m 100%\u001b[0m \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m2,120/2,120 \u001b[0m [ \u001b[33m0:00:38\u001b[0m < \u001b[36m0:00:00\u001b[0m , \u001b[31m56 it/s\u001b[0m ]t/s\u001b[0m ]t/s\u001b[0m ]\n",
            "\u001b[?25h\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m sat_3 - Iteration \u001b[1;36m30\u001b[0m of \u001b[1;36m35\u001b[0m                                            \n",
            "\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m Generating alignments\u001b[33m...\u001b[0m                                              \n",
            "\u001b[2K\u001b[35m 100%\u001b[0m \u001b[91m━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[91m╸\u001b[0m \u001b[32m2,116/2,120 \u001b[0m [ \u001b[33m0:01:03\u001b[0m < \u001b[36m0:00:01\u001b[0m , \u001b[31m35 it/s\u001b[0m ]t/s\u001b[0m ]\n",
            "\u001b[?25h\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m Accumulating statistics\u001b[33m...\u001b[0m                                            \n",
            "\u001b[2K\u001b[35m 100%\u001b[0m \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m2,121/2,120 \u001b[0m [ \u001b[33m0:00:38\u001b[0m < \u001b[36m0:00:00\u001b[0m , \u001b[31m55 it/s\u001b[0m ]t/s\u001b[0m ]t/s\u001b[0m ]\n",
            "\u001b[?25h\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m sat_3 - Iteration \u001b[1;36m31\u001b[0m of \u001b[1;36m35\u001b[0m                                            \n",
            "\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m Accumulating statistics\u001b[33m...\u001b[0m                                            \n",
            "\u001b[2K\u001b[35m 100%\u001b[0m \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m2,127/2,120 \u001b[0m [ \u001b[33m0:00:39\u001b[0m < \u001b[36m0:00:00\u001b[0m , \u001b[31m54 it/s\u001b[0m ]t/s\u001b[0m ]t/s\u001b[0m ]\n",
            "\u001b[?25h\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m sat_3 - Iteration \u001b[1;36m32\u001b[0m of \u001b[1;36m35\u001b[0m                                            \n",
            "\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m Accumulating statistics\u001b[33m...\u001b[0m                                            \n",
            "\u001b[2K\u001b[35m 100%\u001b[0m \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m2,124/2,120 \u001b[0m [ \u001b[33m0:00:39\u001b[0m < \u001b[36m0:00:00\u001b[0m , \u001b[31m53 it/s\u001b[0m ]t/s\u001b[0m ]t/s\u001b[0m ]\n",
            "\u001b[?25h\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m sat_3 - Iteration \u001b[1;36m33\u001b[0m of \u001b[1;36m35\u001b[0m                                            \n",
            "\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m Accumulating statistics\u001b[33m...\u001b[0m                                            \n",
            "\u001b[2K\u001b[35m 100%\u001b[0m \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m2,124/2,120 \u001b[0m [ \u001b[33m0:00:39\u001b[0m < \u001b[36m0:00:00\u001b[0m , \u001b[31m54 it/s\u001b[0m ]t/s\u001b[0m ]t/s\u001b[0m ]\n",
            "\u001b[?25h\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m sat_3 - Iteration \u001b[1;36m34\u001b[0m of \u001b[1;36m35\u001b[0m                                            \n",
            "\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m Accumulating statistics\u001b[33m...\u001b[0m                                            \n",
            "\u001b[2K\u001b[35m 100%\u001b[0m \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m2,128/2,120 \u001b[0m [ \u001b[33m0:00:39\u001b[0m < \u001b[36m0:00:00\u001b[0m , \u001b[31m54 it/s\u001b[0m ]t/s\u001b[0m ]t/s\u001b[0m ]\n",
            "\u001b[?25h\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m sat_3 - Iteration \u001b[1;36m35\u001b[0m of \u001b[1;36m35\u001b[0m                                            \n",
            "\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m Accumulating statistics\u001b[33m...\u001b[0m                                            \n",
            "\u001b[2K\u001b[35m 100%\u001b[0m \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m2,124/2,120 \u001b[0m [ \u001b[33m0:00:39\u001b[0m < \u001b[36m0:00:00\u001b[0m , \u001b[31m54 it/s\u001b[0m ]t/s\u001b[0m ]t/s\u001b[0m ]\n",
            "\u001b[?25h\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m Creating alignment model for speaker-independent features\u001b[33m...\u001b[0m          \n",
            "\u001b[2K\u001b[35m 100%\u001b[0m \u001b[91m━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[91m╸\u001b[0m \u001b[32m2,119/2,120 \u001b[0m [ \u001b[33m0:00:40\u001b[0m < \u001b[36m0:00:01\u001b[0m , \u001b[31m55 it/s\u001b[0m ]t/s\u001b[0m ]\n",
            "\u001b[?25h\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m Training complete!                                                    \n",
            "\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m Exiting training early to save time as the corpus is below the subset \n",
            "\u001b[2;36m \u001b[0m         size for later training stages                                        \n",
            "\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m Compiling training graphs\u001b[33m...\u001b[0m                                          \n",
            "\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m Generating alignments\u001b[33m...\u001b[0m                                              \n",
            "\u001b[2K\u001b[35m 100%\u001b[0m \u001b[91m━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[91m╸\u001b[0m \u001b[32m2,116/2,120 \u001b[0m [ \u001b[33m0:01:03\u001b[0m < \u001b[36m0:00:01\u001b[0m , \u001b[31m35 it/s\u001b[0m ]t/s\u001b[0m ]\n",
            "\u001b[?25h\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m Calculating fMLLR for speaker adaptation\u001b[33m...\u001b[0m                           \n",
            "\u001b[2K\u001b[35m 100%\u001b[0m \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m12/1 \u001b[0m [ \u001b[33m0:00:04\u001b[0m < \u001b[36m0:00:00\u001b[0m , \u001b[31m11 it/s\u001b[0m ]\n",
            "\u001b[?25h\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m Generating alignments\u001b[33m...\u001b[0m                                              \n",
            "\u001b[2K\u001b[35m 100%\u001b[0m \u001b[91m━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[91m╸\u001b[0m \u001b[32m2,113/2,120 \u001b[0m [ \u001b[33m0:01:03\u001b[0m < \u001b[36m0:00:01\u001b[0m , \u001b[31m35 it/s\u001b[0m ]t/s\u001b[0m ]\n",
            "\u001b[?25h\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m Accumulating transition stats\u001b[33m...\u001b[0m                                      \n",
            "\u001b[2K\u001b[35m  95%\u001b[0m \u001b[91m━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[91m╸\u001b[0m\u001b[90m━\u001b[0m \u001b[32m2,008/2,120 \u001b[0m [ \u001b[33m0:00:03\u001b[0m < \u001b[36m0:00:01\u001b[0m , \u001b[31m1,505 it/s\u001b[0m ]\n",
            "\u001b[?25h\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m Finished accumulating transition stats!                               \n",
            "\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m Collecting phone and word alignments from sat_3_ali lattices\u001b[33m...\u001b[0m       \n",
            "\u001b[2K\u001b[35m  94%\u001b[0m \u001b[91m━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━\u001b[0m \u001b[32m1,984/2,120 \u001b[0m [ \u001b[33m0:00:06\u001b[0m < \u001b[36m0:00:01\u001b[0m , \u001b[31m3,300 it/s\u001b[0m ]\n",
            "\u001b[?25h\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m Beginning phone LM training\u001b[33m...\u001b[0m                                        \n",
            "\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m Collecting training data\u001b[33m...\u001b[0m                                           \n",
            "\u001b[2K\u001b[35m   0%\u001b[0m \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m0/2,120 \u001b[0m [ \u001b[33m0:00:01\u001b[0m < \u001b[36m-:--:--\u001b[0m , \u001b[31m? it/s\u001b[0m ]\n",
            "\u001b[?25h\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m Training model\u001b[33m...\u001b[0m                                                     \n",
            "\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m Completed training in \u001b[1;36m2789.003955602646\u001b[0m seconds!                      \n",
            "\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m Saved model to audio_1_experiment/audio_1_ver2_mfa                    \n",
            "\u001b[2;36m \u001b[0m\u001b[33mWARNING \u001b[0m Alignment analysis not available without using postgresql             \n",
            "\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m Exporting sat_3_ali TextGrids to audio_1_experiment/audio_1_ver2\u001b[33m...\u001b[0m   \n",
            "\u001b[2K\u001b[35m  99%\u001b[0m \u001b[91m━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[91m╸\u001b[0m \u001b[32m2,109/2,120 \u001b[0m [ \u001b[33m0:00:16\u001b[0m < \u001b[36m0:00:01\u001b[0m , \u001b[31m149 it/s\u001b[0m ]t/s\u001b[0m ]\n",
            "\u001b[?25h\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m Finished exporting TextGrids to audio_1_experiment/audio_1_ver2!      \n",
            "\u001b[2;36m \u001b[0m\u001b[32mINFO    \u001b[0m Done! Everything took \u001b[1;36m2821.442\u001b[0m seconds                                \n"
          ]
        }
      ],
      "source": [
        "!source miniconda/bin/activate aligner; \\\n",
        "mfa train \\\n",
        "    --num_jobs `nproc` \\\n",
        "    --use_mp \\\n",
        "    --clean \\\n",
        "    --overwrite \\\n",
        "    --no_textgrid_cleanup \\\n",
        "    --single_speaker \\\n",
        "    --output_format json \\\n",
        "    --output_directory audio_1_experiment/audio_1_ver2 \\\n",
        "    audio_1_experiment/audio_1_ver2 ./lexicon.txt audio_1_experiment/audio_1_ver2_mfa"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "D8s22KjOwtew"
      },
      "outputs": [],
      "source": []
    },
    {
      "cell_type": "code",
      "execution_count": 25,
      "metadata": {
        "id": "uy88PuP4TtoU"
      },
      "outputs": [],
      "source": [
        "#### PREPARE TF DATASET ####\n",
        "# Note: We are using a sampling rate of 16k,\n",
        "# even though the original data has a sampling rate of 48k."
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 26,
      "metadata": {
        "id": "p58RNMt8CISF"
      },
      "outputs": [
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "/home/billgates/miniconda3/envs/TTS_experiment_3/lib/python3.7/site-packages/tqdm/auto.py:21: TqdmWarning: IProgress not found. Please update jupyter and ipywidgets. See https://ipywidgets.readthedocs.io/en/stable/user_install.html\n",
            "  from .autonotebook import tqdm as notebook_tqdm\n",
            "2024-01-25 10:55:12.112627: I tensorflow/core/platform/cpu_feature_guard.cc:193] This TensorFlow binary is optimized with oneAPI Deep Neural Network Library (oneDNN) to use the following CPU instructions in performance-critical operations:  AVX2 FMA\n",
            "To enable them in other operations, rebuild TensorFlow with the appropriate compiler flags.\n",
            "2024-01-25 10:55:12.790540: W tensorflow/compiler/xla/stream_executor/platform/default/dso_loader.cc:64] Could not load dynamic library 'libnvinfer.so.7'; dlerror: libnvinfer.so.7: cannot open shared object file: No such file or directory\n",
            "2024-01-25 10:55:12.790604: W tensorflow/compiler/xla/stream_executor/platform/default/dso_loader.cc:64] Could not load dynamic library 'libnvinfer_plugin.so.7'; dlerror: libnvinfer_plugin.so.7: cannot open shared object file: No such file or directory\n",
            "2024-01-25 10:55:12.790610: W tensorflow/compiler/tf2tensorrt/utils/py_utils.cc:38] TF-TRT Warning: Cannot dlopen some TensorRT libraries. If you would like to use Nvidia GPU with TensorRT, please make sure the missing libraries mentioned above are installed properly.\n"
          ]
        }
      ],
      "source": [
        "import json\n",
        "from pathlib import Path\n",
        "import numpy as np\n",
        "import torch\n",
        "import json\n",
        "import librosa\n",
        "import tensorflow as tf\n",
        "from tqdm.auto import tqdm\n",
        "import random"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 27,
      "metadata": {},
      "outputs": [],
      "source": [
        "import json\n",
        "from pathlib import Path\n",
        "import numpy as np\n",
        "import torch\n",
        "import json\n",
        "import librosa\n",
        "import tensorflow as tf\n",
        "from tqdm.auto import tqdm\n",
        "import random"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 40,
      "metadata": {
        "id": "YbF8kuar0zuC"
      },
      "outputs": [],
      "source": [
        "mel_basis = {}\n",
        "hann_window = {}\n",
        "\n",
        "def spectrogram_torch(y, n_fft, sampling_rate, hop_size, win_size, center=False):\n",
        "    global hann_window\n",
        "    # if torch.min(y) < -1.0:\n",
        "    #     print(\"min value is \", torch.min(y))\n",
        "    # if torch.max(y) > 1.0:\n",
        "    #     print(\"max value is \", torch.max(y))\n",
        "\n",
        "    dtype_device = str(y.dtype) + \"_\" + str(y.device)\n",
        "    wnsize_dtype_device = str(win_size) + \"_\" + dtype_device\n",
        "    if wnsize_dtype_device not in hann_window:\n",
        "        hann_window[wnsize_dtype_device] = torch.hann_window(win_size).to(\n",
        "            dtype=y.dtype, device=y.device\n",
        "        )\n",
        "\n",
        "    y = torch.nn.functional.pad(\n",
        "        y.unsqueeze(1),\n",
        "        (int((n_fft - hop_size) / 2), int((n_fft - hop_size) / 2)),\n",
        "        mode=\"reflect\",\n",
        "    )\n",
        "    y = y.squeeze(1)\n",
        "\n",
        "    spec = torch.stft(\n",
        "        y,\n",
        "        n_fft,\n",
        "        hop_length=hop_size,\n",
        "        win_length=win_size,\n",
        "        window=hann_window[wnsize_dtype_device],\n",
        "        center=center,\n",
        "        pad_mode=\"reflect\",\n",
        "        normalized=False,\n",
        "        onesided=True,\n",
        "        return_complex=True\n",
        "    )\n",
        "    spec = torch.view_as_real(spec)\n",
        "    spec = torch.sqrt(spec.pow(2).sum(-1) + 1e-6)\n",
        "    spec = spec.squeeze(0)\n",
        "    return torch.swapaxes(spec, 0, 1)\n",
        "\n",
        "\n",
        "def tensor_to_bytes(t):\n",
        "    t = tf.constant(t)\n",
        "    t = tf.io.serialize_tensor(t)\n",
        "    return tf.train.Feature(bytes_list=tf.train.BytesList(value=[t.numpy()]))\n",
        "\n",
        "\n",
        "def write_tfdata(data, out_file):\n",
        "    with tf.io.TFRecordWriter(out_file) as file_writer:\n",
        "        for wav_file, seq, _ in data:\n",
        "            phone_seq = []\n",
        "            for phone, duration in seq:\n",
        "                phone_idx = phone_set.index(phone)\n",
        "                phone_seq.append((phone_idx, duration))\n",
        "            phone_seq = np.array(phone_seq, dtype=np.float32)\n",
        "\n",
        "            # load wav\n",
        "            wav, sr = librosa.load(wav_file, sr=config[\"data\"][\"sampling_rate\"], dtype=np.float32)\n",
        "            wav = torch.from_numpy(wav)\n",
        "            # compute spec\n",
        "            spec = spectrogram_torch(\n",
        "                wav[None],\n",
        "                n_fft=config[\"data\"][\"filter_length\"],\n",
        "                sampling_rate=config[\"data\"][\"sampling_rate\"],\n",
        "                hop_size=config[\"data\"][\"hop_length\"],\n",
        "                win_size=config[\"data\"][\"win_length\"],\n",
        "                center=False\n",
        "            )\n",
        "            features = {\n",
        "                \"phone_idx\": tensor_to_bytes(phone_seq[:, 0].astype(np.int32)),\n",
        "                \"phone_duration\": tensor_to_bytes(phone_seq[:, 1]),\n",
        "                \"wav\": tensor_to_bytes(wav.half().numpy()),\n",
        "                \"spec\": tensor_to_bytes(spec.half().numpy())\n",
        "            }\n",
        "            example = tf.train.Example(features=tf.train.Features(feature=features))\n",
        "            file_writer.write(example.SerializeToString())\n",
        "\n",
        "def write_split(split, data, num_chunks):\n",
        "    data = np.array(data, dtype=object)\n",
        "    chunks = list(np.array_split(data, num_chunks))\n",
        "    for i, chunk in enumerate(tqdm(chunks)):\n",
        "        write_tfdata(chunk, f\"tfdata_audio_1/{split}/part_{i:03d}.tfrecords\")"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 41,
      "metadata": {
        "id": "UIAxgVx-DkRI"
      },
      "outputs": [],
      "source": [
        "!mkdir -p tfdata_audio_1/{train,test}"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 42,
      "metadata": {},
      "outputs": [
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "[Errno 2] No such file or directory: 'data'\n",
            "/home/billgates/lightspeed/data\n"
          ]
        }
      ],
      "source": [
        "%cd data"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 43,
      "metadata": {
        "id": "STURjAulDm5B"
      },
      "outputs": [],
      "source": [
        "with open(\"../config.json\", \"rb\") as f:\n",
        "    config = json.load(f)\n",
        "device = \"cuda\" if torch.cuda.is_available() else \"cpu\"\n",
        "data_dir = Path(\"audio_1_experiment/audio_1_ver2/\")\n",
        "json_files = sorted(data_dir.glob(\"*.json\"))\n",
        "dataset = []\n",
        "phone_set = []\n",
        "\n",
        "for file_path in json_files:\n",
        "    with open(file_path, \"rb\") as f:\n",
        "        data = json.load(f)\n",
        "    seq = []\n",
        "    word_index = 0\n",
        "    words =  data[\"tiers\"][\"words\"][\"entries\"]\n",
        "    for start, end, phone in data[\"tiers\"][\"phones\"][\"entries\"]:\n",
        "        if start > words[word_index][1] - 1e-5:\n",
        "            seq.append( (\"<SEP>\", 0) )\n",
        "            word_index += 1\n",
        "        duration = end * 1000 - start * 1000 # ms\n",
        "        phone_set.append(phone)\n",
        "        seq.append( (phone, duration) )\n",
        "    # print(\"seq\", seq)\n",
        "    wav_file = file_path.with_suffix(\".wav\")\n",
        "    dataset.append((wav_file, seq, data[\"end\"]))\n",
        "\n",
        "phone_set = [\"<SEP>\"] + sorted(set(phone_set))\n",
        "assert len(phone_set) <= 256\n",
        "with open(\"phone_set.json\", \"w\", encoding=\"utf-8\") as f:\n",
        "    json.dump(phone_set, f)\n",
        "\n",
        "assert phone_set.index(\"<SEP>\") == 0"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 44,
      "metadata": {
        "id": "2d86zvZBD0r6"
      },
      "outputs": [
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "Train data size: 1860\n",
            "Test data size: 256\n"
          ]
        }
      ],
      "source": [
        "random.Random(42).shuffle(dataset)\n",
        "L = len(dataset) - 256\n",
        "train_data = dataset[:L]\n",
        "test_data = dataset[L:]\n",
        "print(\"Train data size:\", len(train_data))\n",
        "print(\"Test data size:\", len(test_data))"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 45,
      "metadata": {},
      "outputs": [
        {
          "data": {
            "text/plain": [
              "'/home/billgates/lightspeed/data'"
            ]
          },
          "execution_count": 45,
          "metadata": {},
          "output_type": "execute_result"
        }
      ],
      "source": [
        "%pwd"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 46,
      "metadata": {},
      "outputs": [
        {
          "data": {
            "text/plain": [
              "(PosixPath('audio_1_experiment/audio_1_ver2/audio_1-5827.wav'),\n",
              " [('c', 10.0),\n",
              "  ('ả', 10.0),\n",
              "  ('m', 10.0),\n",
              "  ('<SEP>', 0),\n",
              "  ('t', 20.0),\n",
              "  ('h', 10.0),\n",
              "  ('ấ', 10.0),\n",
              "  ('y', 50.0),\n",
              "  ('<SEP>', 0),\n",
              "  ('sil', 80.0),\n",
              "  ('<SEP>', 0),\n",
              "  ('g', 10.0),\n",
              "  ('h', 10.0),\n",
              "  ('e', 90.0),\n",
              "  ('n', 120.0),\n",
              "  ('<SEP>', 0),\n",
              "  ('t', 60.0),\n",
              "  ('ị', 180.0),\n",
              "  ('<SEP>', 0),\n",
              "  ('v', 60.0),\n",
              "  ('ớ', 50.0),\n",
              "  ('i', 60.0),\n",
              "  ('<SEP>', 0),\n",
              "  ('h', 130.0),\n",
              "  ('ọ', 320.0),\n",
              "  ('<SEP>', 0),\n",
              "  ('sil', 100.0),\n",
              "  ('<SEP>', 0),\n",
              "  ('h', 100.0),\n",
              "  ('ỏ', 40.0),\n",
              "  ('i', 30.0),\n",
              "  ('<SEP>', 0),\n",
              "  ('sil', 30.0),\n",
              "  ('<SEP>', 0),\n",
              "  ('đ', 50.0),\n",
              "  ('ộ', 140.0),\n",
              "  ('c', 20.0),\n",
              "  ('<SEP>', 0),\n",
              "  ('l', 90.0),\n",
              "  ('ậ', 60.0),\n",
              "  ('p', 20.0),\n",
              "  ('<SEP>', 0),\n",
              "  ('sil', 210.0),\n",
              "  ('<SEP>', 0),\n",
              "  ('t', 60.0),\n",
              "  ('ự', 180.0),\n",
              "  ('<SEP>', 0),\n",
              "  ('t', 40.0),\n",
              "  ('i', 70.0),\n",
              "  ('n', 130.0),\n",
              "  ('<SEP>', 0),\n",
              "  ('sil', 30.0),\n",
              "  ('<SEP>', 0),\n",
              "  ('v', 70.0),\n",
              "  ('à', 10.0),\n",
              "  ('<SEP>', 0),\n",
              "  ('sil', 60.0),\n",
              "  ('<SEP>', 0),\n",
              "  ('đ', 40.0),\n",
              "  ('ư', 70.0),\n",
              "  ('ờ', 10.0),\n",
              "  ('n', 20.0),\n",
              "  ('g', 10.0),\n",
              "  ('<SEP>', 0),\n",
              "  ('sil', 70.0),\n",
              "  ('<SEP>', 0),\n",
              "  ('h', 90.0),\n",
              "  ('o', 130.0),\n",
              "  ('à', 10.0),\n",
              "  ('n', 20.0),\n",
              "  ('g', 10.0),\n",
              "  ('<SEP>', 0),\n",
              "  ('sil', 510.0),\n",
              "  ('<SEP>', 0),\n",
              "  ('t', 20.0),\n",
              "  ('r', 40.0),\n",
              "  ('o', 40.0),\n",
              "  ('n', 50.0),\n",
              "  ('g', 10.0),\n",
              "  ('<SEP>', 0),\n",
              "  ('k', 20.0),\n",
              "  ('h', 40.0),\n",
              "  ('i', 120.0),\n",
              "  ('<SEP>', 0),\n",
              "  ('sil', 30.0),\n",
              "  ('<SEP>', 0),\n",
              "  ('a', 110.0),\n",
              "  ('n', 170.0),\n",
              "  ('h', 10.0),\n",
              "  ('<SEP>', 0),\n",
              "  ('c', 20.0),\n",
              "  ('h', 50.0),\n",
              "  ('ỉ', 80.0),\n",
              "  ('<SEP>', 0),\n",
              "  ('đ', 60.0),\n",
              "  ('ơ', 100.0),\n",
              "  ('n', 80.0),\n",
              "  ('<SEP>', 0),\n",
              "  ('sil', 398.02700000000004)],\n",
              " 5.268027)"
            ]
          },
          "execution_count": 46,
          "metadata": {},
          "output_type": "execute_result"
        }
      ],
      "source": [
        "train_data[0]"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {},
      "outputs": [],
      "source": []
    },
    {
      "cell_type": "code",
      "execution_count": 47,
      "metadata": {
        "id": "UnDTA1sKD9Ug"
      },
      "outputs": [
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  0%|          | 0/1 [00:00<?, ?it/s]2024-01-25 10:56:34.714317: I tensorflow/compiler/xla/stream_executor/cuda/cuda_gpu_executor.cc:981] successful NUMA node read from SysFS had negative value (-1), but there must be at least one NUMA node, so returning NUMA node zero\n",
            "2024-01-25 10:56:34.714711: W tensorflow/compiler/xla/stream_executor/platform/default/dso_loader.cc:64] Could not load dynamic library 'libcufft.so.10'; dlerror: libcufft.so.10: cannot open shared object file: No such file or directory\n",
            "2024-01-25 10:56:34.714785: W tensorflow/compiler/xla/stream_executor/platform/default/dso_loader.cc:64] Could not load dynamic library 'libcurand.so.10'; dlerror: libcurand.so.10: cannot open shared object file: No such file or directory\n",
            "2024-01-25 10:56:34.714851: W tensorflow/compiler/xla/stream_executor/platform/default/dso_loader.cc:64] Could not load dynamic library 'libcusolver.so.11'; dlerror: libcusolver.so.11: cannot open shared object file: No such file or directory\n",
            "2024-01-25 10:56:34.714912: W tensorflow/compiler/xla/stream_executor/platform/default/dso_loader.cc:64] Could not load dynamic library 'libcusparse.so.11'; dlerror: libcusparse.so.11: cannot open shared object file: No such file or directory\n",
            "2024-01-25 10:56:34.714940: W tensorflow/core/common_runtime/gpu/gpu_device.cc:1934] Cannot dlopen some GPU libraries. Please make sure the missing libraries mentioned above are installed properly if you would like to use GPU. Follow the guide at https://www.tensorflow.org/install/gpu for how to download and setup the required libraries for your platform.\n",
            "Skipping registering GPU devices...\n",
            "2024-01-25 10:56:34.716209: I tensorflow/core/platform/cpu_feature_guard.cc:193] This TensorFlow binary is optimized with oneAPI Deep Neural Network Library (oneDNN) to use the following CPU instructions in performance-critical operations:  AVX2 FMA\n",
            "To enable them in other operations, rebuild TensorFlow with the appropriate compiler flags.\n",
            "100%|██████████| 1/1 [00:02<00:00,  2.40s/it]\n"
          ]
        }
      ],
      "source": [
        "write_split(\"test\", test_data, 1)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 48,
      "metadata": {
        "id": "JdG7lqdEEQNz"
      },
      "outputs": [
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "100%|██████████| 256/256 [00:08<00:00, 30.09it/s]\n"
          ]
        }
      ],
      "source": [
        "write_split(\"train\", train_data, 256)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "M7ZIyB_ANeSa"
      },
      "outputs": [],
      "source": []
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {},
      "outputs": [],
      "source": []
    }
  ],
  "metadata": {
    "accelerator": "TPU",
    "colab": {
      "machine_shape": "hm",
      "provenance": []
    },
    "kernelspec": {
      "display_name": "Python 3",
      "name": "python3"
    },
    "language_info": {
      "codemirror_mode": {
        "name": "ipython",
        "version": 3
      },
      "file_extension": ".py",
      "mimetype": "text/x-python",
      "name": "python",
      "nbconvert_exporter": "python",
      "pygments_lexer": "ipython3",
      "version": "3.7.12"
    }
  },
  "nbformat": 4,
  "nbformat_minor": 0
}