windows conda 环境 tensorrt问题

[kawae0709]

python webui.py --port 50000 --model_dir pretrained_models/CosyVoice2-0.5B

2025-01-12 22:55:20,338 - modelscope - INFO - PyTorch version 2.3.1+cu121 Found.
2025-01-12 22:55:20,342 - modelscope - INFO - Loading ast index from C:\Users\kawae.cache\modelscope\ast_indexer
2025-01-12 22:55:21,549 - modelscope - INFO - Loading done! Current index file version is 1.15.0, with md5 4055a8dd19698e8003c68c14549359b5 and a total number of 980 components indexed
failed to import ttsfrd, use WeTextProcessing instead
Traceback (most recent call last):
File "D:\AIgithub\CosyVoice\webui.py", line 25, in
from cosyvoice.cli.cosyvoice import CosyVoice, CosyVoice2
File "D:\AIgithub\CosyVoice\cosyvoice\cli\cosyvoice.py", line 21, in
from cosyvoice.cli.model import CosyVoiceModel, CosyVoice2Model
File "D:\AIgithub\CosyVoice\cosyvoice\cli\model.py", line 23, in
from cosyvoice.utils.file_utils import convert_onnx_to_trt
File "D:\AIgithub\CosyVoice\cosyvoice\utils\file_utils.py", line 17, in
import tensorrt as trt
ModuleNotFoundError: No module named 'tensorrt'

以前是正常的,最近几天的更新,好像必须tensorrt了,win11 conda环境破不了

[aluminumbox]

更新代码，已修复

[kawae0709]

2025-01-14 13:14:33,962 INFO get zero_shot inference request
0%| | 0/12 [00:00<?, ?it/s]2025-01-14 13:14:34,368 INFO synthesis text 有人曾说：你的气质里，藏着你走过的路，读过的书和爱过的人。或许董宇辉就是这样的人，生而平凡没有神的光环，但是只要通过不懈的坚持，也总归能够坦坦荡荡，落落大方。
2025-01-14 13:14:39,148 INFO yield speech len 13.72, rtf 0.34841314696709547
2025-01-14 13:14:39,183 DEBUG subprocess.call(['ffmpeg', '-y', '-read_ahead_limit', '-1', '-i', 'cache:pipe:0', '-acodec', 'pcm_s16le', '-vn', '-f', 'wav', '-'])
2025-01-14 13:14:39,305 DEBUG subprocess.call(['ffmpeg', '-y', '-f', 'wav', '-i', 'C:\Users\kawae\AppData\Local\Temp\tmpm0we9alt', '-f', 'adts', 'C:\Users\kawae\AppData\Local\Temp\tmpnrtxngtq'])
2025-01-14 13:14:39,377 DEBUG subprocess output: b'ffmpeg version 7.1 Copyright (c) 2000-2024 the FFmpeg developers'
2025-01-14 13:14:39,377 DEBUG subprocess output: b' built with clang version 19.1.0'
2025-01-14 13:14:39,377 DEBUG subprocess output: b' configuration: --prefix=/d/bld/ffmpeg_1728332459246/_h_env/Library --cc=clang.exe --cxx=clang++.exe --nm=llvm-nm --ar=llvm-ar --disable-doc --enable-openssl --enable-demuxer=dash --enable-hardcoded-tables --enable-libfreetype --enable-libharfbuzz --enable-libfontconfig --enable-libopenh264 --enable-libdav1d --ld=lld-link --target-os=win64 --enable-cross-compile --toolchain=msvc --host-cc=clang.exe --extra-libs=ucrt.lib --extra-libs=vcruntime.lib --extra-libs=oldnames.lib --strip=llvm-strip --disable-stripping --host-extralibs= --disable-libopenvino --enable-gpl --enable-libx264 --enable-libx265 --enable-libaom --enable-libsvtav1 --enable-libxml2 --enable-pic --enable-shared --disable-static --enable-version3 --enable-zlib --enable-libopus --pkg-config=/d/bld/ffmpeg_1728332459246/_build_env/Library/bin/pkg-config'
2025-01-14 13:14:39,377 DEBUG subprocess output: b' libavutil 59. 39.100 / 59. 39.100'
2025-01-14 13:14:39,377 DEBUG subprocess output: b' libavcodec 61. 19.100 / 61. 19.100'
2025-01-14 13:14:39,378 DEBUG subprocess output: b' libavformat 61. 7.100 / 61. 7.100'
2025-01-14 13:14:39,378 DEBUG subprocess output: b' libavdevice 61. 3.100 / 61. 3.100'
2025-01-14 13:14:39,378 DEBUG subprocess output: b' libavfilter 10. 4.100 / 10. 4.100'
2025-01-14 13:14:39,378 DEBUG subprocess output: b' libswscale 8. 3.100 / 8. 3.100'
2025-01-14 13:14:39,378 DEBUG subprocess output: b' libswresample 5. 3.100 / 5. 3.100'
2025-01-14 13:14:39,378 DEBUG subprocess output: b' libpostproc 58. 3.100 / 58. 3.100'
2025-01-14 13:14:39,378 DEBUG subprocess output: b'[aist#0:0/pcm_s16le @ 0000028AE08684C0] Guessed Channel Layout: mono'
2025-01-14 13:14:39,378 DEBUG subprocess output: b"Input #0, wav, from 'C:\Users\kawae\AppData\Local\Temp\tmpm0we9alt':"
2025-01-14 13:14:39,378 DEBUG subprocess output: b' Duration: 00:00:13.72, bitrate: 384 kb/s'
2025-01-14 13:14:39,378 DEBUG subprocess output: b' Stream #0:0: Audio: pcm_s16le ([1][0][0][0] / 0x0001), 24000 Hz, mono, s16, 384 kb/s'
2025-01-14 13:14:39,378 DEBUG subprocess output: b'Stream mapping:'
2025-01-14 13:14:39,378 DEBUG subprocess output: b' Stream #0:0 -> #0:0 (pcm_s16le (native) -> aac (native))'
2025-01-14 13:14:39,378 DEBUG subprocess output: b'Press [q] to stop, [?] for help'
2025-01-14 13:14:39,378 DEBUG subprocess output: b"Output #0, adts, to 'C:\Users\kawae\AppData\Local\Temp\tmpnrtxngtq':"
2025-01-14 13:14:39,378 DEBUG subprocess output: b' Metadata:'
2025-01-14 13:14:39,378 DEBUG subprocess output: b' encoder : Lavf61.7.100'
2025-01-14 13:14:39,378 DEBUG subprocess output: b' Stream #0:0: Audio: aac (LC), 24000 Hz, mono, fltp, 69 kb/s'
2025-01-14 13:14:39,378 DEBUG subprocess output: b' Metadata:'
2025-01-14 13:14:39,378 DEBUG subprocess output: b' encoder : Lavc61.19.100 aac'
2025-01-14 13:14:39,378 DEBUG subprocess output: b'[out#0/adts @ 0000028AE089ADC0] video:0KiB audio:122KiB subtitle:0KiB other streams:0KiB global headers:0KiB muxing overhead: 1.804815%'
2025-01-14 13:14:39,378 DEBUG subprocess output: b'size= 125KiB time=00:00:13.72 bitrate= 74.4kbits/s speed= 256x'
2025-01-14 13:14:39,378 DEBUG subprocess output: b'[aac @ 0000028AE08C1500] Qavg: 400.077'
8%|████████ | 1/12 [00:05<00:58, 5.30s/it]2025-01-14 13:14:39,664 INFO synthesis text 今天给大家整理了一些董宇辉的金句，希望我们都能学会表达，闪闪发光。当你背单词时，阿拉斯加的鳕鱼正跃出水面；当你算数学时，南太平洋的海鸥正掠过海岸；
2025-01-14 13:14:44,299 INFO yield speech len 14.24, rtf 0.32548554492800424
2025-01-14 13:14:44,333 DEBUG subprocess.call(['ffmpeg', '-y', '-read_ahead_limit', '-1', '-i', 'cache:pipe:0', '-acodec', 'pcm_s16le', '-vn', '-f', 'wav', '-'])
2025-01-14 13:14:44,354 DEBUG subprocess.call(['ffmpeg', '-y', '-f', 'wav', '-i', 'C:\Users\kawae\AppData\Local\Temp\tmpvev_bx3n', '-f', 'adts', 'C:\Users\kawae\AppData\Local\Temp\tmptxh6gz8i'])
2025-01-14 13:14:44,442 DEBUG subprocess output: b'ffmpeg version 7.1 Copyright (c) 2000-2024 the FFmpeg developers'
2025-01-14 13:14:44,442 DEBUG subprocess output: b' built with clang version 19.1.0'
2025-01-14 13:14:44,443 DEBUG subprocess output: b' configuration: --prefix=/d/bld/ffmpeg_1728332459246/_h_env/Library --cc=clang.exe --cxx=clang++.exe --nm=llvm-nm --ar=llvm-ar --disable-doc --enable-openssl --enable-demuxer=dash --enable-hardcoded-tables --enable-libfreetype --enable-libharfbuzz --enable-libfontconfig --enable-libopenh264 --enable-libdav1d --ld=lld-link --target-os=win64 --enable-cross-compile --toolchain=msvc --host-cc=clang.exe --extra-libs=ucrt.lib --extra-libs=vcruntime.lib --extra-libs=oldnames.lib --strip=llvm-strip --disable-stripping --host-extralibs= --disable-libopenvino --enable-gpl --enable-libx264 --enable-libx265 --enable-libaom --enable-libsvtav1 --enable-libxml2 --enable-pic --enable-shared --disable-static --enable-version3 --enable-zlib --enable-libopus --pkg-config=/d/bld/ffmpeg_1728332459246/_build_env/Library/bin/pkg-config'
2025-01-14 13:14:44,443 DEBUG subprocess output: b' libavutil 59. 39.100 / 59. 39.100'
2025-01-14 13:14:44,443 DEBUG subprocess output: b' libavcodec 61. 19.100 / 61. 19.100'
2025-01-14 13:14:44,443 DEBUG subprocess output: b' libavformat 61. 7.100 / 61. 7.100'
2025-01-14 13:14:44,443 DEBUG subprocess output: b' libavdevice 61. 3.100 / 61. 3.100'
2025-01-14 13:14:44,443 DEBUG subprocess output: b' libavfilter 10. 4.100 / 10. 4.100'
2025-01-14 13:14:44,443 DEBUG subprocess output: b' libswscale 8. 3.100 / 8. 3.100'
2025-01-14 13:14:44,443 DEBUG subprocess output: b' libswresample 5. 3.100 / 5. 3.100'
2025-01-14 13:14:44,443 DEBUG subprocess output: b' libpostproc 58. 3.100 / 58. 3.100'
2025-01-14 13:14:44,443 DEBUG subprocess output: b'[aist#0:0/pcm_s16le @ 000001F79FFE42C0] Guessed Channel Layout: mono'
2025-01-14 13:14:44,443 DEBUG subprocess output: b"Input #0, wav, from 'C:\Users\kawae\AppData\Local\Temp\tmpvev_bx3n':"
2025-01-14 13:14:44,443 DEBUG subprocess output: b' Duration: 00:00:14.24, bitrate: 384 kb/s'
2025-01-14 13:14:44,443 DEBUG subprocess output: b' Stream #0:0: Audio: pcm_s16le ([1][0][0][0] / 0x0001), 24000 Hz, mono, s16, 384 kb/s'
2025-01-14 13:14:44,443 DEBUG subprocess output: b'Stream mapping:'
2025-01-14 13:14:44,443 DEBUG subprocess output: b' Stream #0:0 -> #0:0 (pcm_s16le (native) -> aac (native))'
2025-01-14 13:14:44,443 DEBUG subprocess output: b'Press [q] to stop, [?] for help'
2025-01-14 13:14:44,443 DEBUG subprocess output: b"Output #0, adts, to 'C:\Users\kawae\AppData\Local\Temp\tmptxh6gz8i':"
2025-01-14 13:14:44,443 DEBUG subprocess output: b' Metadata:'
2025-01-14 13:14:44,443 DEBUG subprocess output: b' encoder : Lavf61.7.100'
2025-01-14 13:14:44,443 DEBUG subprocess output: b' Stream #0:0: Audio: aac (LC), 24000 Hz, mono, fltp, 69 kb/s'
2025-01-14 13:14:44,443 DEBUG subprocess output: b' Metadata:'
2025-01-14 13:14:44,443 DEBUG subprocess output: b' encoder : Lavc61.19.100 aac'
2025-01-14 13:14:44,443 DEBUG subprocess output: b'[out#0/adts @ 000001F79FFF9100] video:0KiB audio:131KiB subtitle:0KiB other streams:0KiB global headers:0KiB muxing overhead: 1.744247%'
2025-01-14 13:14:44,443 DEBUG subprocess output: b'size= 134KiB time=00:00:14.24 bitrate= 76.8kbits/s speed= 206x'
2025-01-14 13:14:44,443 DEBUG subprocess output: b'[aac @ 000001F7A0026900] Qavg: 982.323'
17%|████████████████ | 2/12 [00:10<00:51, 5.16s/it]2025-01-14 13:14:44,730 INFO synthesis text 当你晚自习时，地球的极圈正五彩斑斓；但少年，梦要你亲自实现，世界你要亲自去看。未来可期，拼尽全力！当你为未来付出踏踏实实努力的时候，那些你觉得看不到的人和遇不到的风景都终将在你的生命里出现。
2025-01-14 13:14:50,730 INFO yield speech len 17.72, rtf 0.3385955001107726
2025-01-14 13:14:50,764 DEBUG subprocess.call(['ffmpeg', '-y', '-read_ahead_limit', '-1', '-i', 'cache:pipe:0', '-acodec', 'pcm_s16le', '-vn', '-f', 'wav', '-'])
2025-01-14 13:14:50,787 DEBUG subprocess.call(['ffmpeg', '-y', '-f', 'wav', '-i', 'C:\Users\kawae\AppData\Local\Temp\tmpqbm9mcyf', '-f', 'adts', 'C:\Users\kawae\AppData\Local\Temp\tmphts7pwc4'])
2025-01-14 13:14:51,015 DEBUG subprocess output: b'ffmpeg version 7.1 Copyright (c) 2000-2024 the FFmpeg developers'
2025-01-14 13:14:51,015 DEBUG subprocess output: b' built with clang version 19.1.0'
2025-01-14 13:14:51,015 DEBUG subprocess output: b' configuration: --prefix=/d/bld/ffmpeg_1728332459246/_h_env/Library --cc=clang.exe --cxx=clang++.exe --nm=llvm-nm --ar=llvm-ar --disable-doc --enable-openssl --enable-demuxer=dash --enable-hardcoded-tables --enable-libfreetype --enable-libharfbuzz --enable-libfontconfig --enable-libopenh264 --enable-libdav1d --ld=lld-link --target-os=win64 --enable-cross-compile --toolchain=msvc --host-cc=clang.exe --extra-libs=ucrt.lib --extra-libs=vcruntime.lib --extra-libs=oldnames.lib --strip=llvm-strip --disable-stripping --host-extralibs= --disable-libopenvino --enable-gpl --enable-libx264 --enable-libx265 --enable-libaom --enable-libsvtav1 --enable-libxml2 --enable-pic --enable-shared --disable-static --enable-version3 --enable-zlib --enable-libopus --pkg-config=/d/bld/ffmpeg_1728332459246/_build_env/Library/bin/pkg-config'
2025-01-14 13:14:51,016 DEBUG subprocess output: b' libavutil 59. 39.100 / 59. 39.100'
2025-01-14 13:14:51,016 DEBUG subprocess output: b' libavcodec 61. 19.100 / 61. 19.100'
2025-01-14 13:14:51,016 DEBUG subprocess output: b' libavformat 61. 7.100 / 61. 7.100'
2025-01-14 13:14:51,016 DEBUG subprocess output: b' libavdevice 61. 3.100 / 61. 3.100'
2025-01-14 13:14:51,016 DEBUG subprocess output: b' libavfilter 10. 4.100 / 10. 4.100'
2025-01-14 13:14:51,016 DEBUG subprocess output: b' libswscale 8. 3.100 / 8. 3.100'
2025-01-14 13:14:51,016 DEBUG subprocess output: b' libswresample 5. 3.100 / 5. 3.100'
2025-01-14 13:14:51,016 DEBUG subprocess output: b' libpostproc 58. 3.100 / 58. 3.100'
2025-01-14 13:14:51,016 DEBUG subprocess output: b'[aist#0:0/pcm_s16le @ 0000021DBE24ABC0] Guessed Channel Layout: mono'
2025-01-14 13:14:51,016 DEBUG subprocess output: b"Input #0, wav, from 'C:\Users\kawae\AppData\Local\Temp\tmpqbm9mcyf':"
2025-01-14 13:14:51,016 DEBUG subprocess output: b' Duration: 00:00:17.72, bitrate: 384 kb/s'
2025-01-14 13:14:51,016 DEBUG subprocess output: b' Stream #0:0: Audio: pcm_s16le ([1][0][0][0] / 0x0001), 24000 Hz, mono, s16, 384 kb/s'
2025-01-14 13:14:51,016 DEBUG subprocess output: b'Stream mapping:'
2025-01-14 13:14:51,016 DEBUG subprocess output: b' Stream #0:0 -> #0:0 (pcm_s16le (native) -> aac (native))'
2025-01-14 13:14:51,016 DEBUG subprocess output: b'Press [q] to stop, [?] for help'
2025-01-14 13:14:51,016 DEBUG subprocess output: b"Output #0, adts, to 'C:\Users\kawae\AppData\Local\Temp\tmphts7pwc4':"
2025-01-14 13:14:51,016 DEBUG subprocess output: b' Metadata:'
2025-01-14 13:14:51,016 DEBUG subprocess output: b' encoder : Lavf61.7.100'
2025-01-14 13:14:51,016 DEBUG subprocess output: b' Stream #0:0: Audio: aac (LC), 24000 Hz, mono, fltp, 69 kb/s'
2025-01-14 13:14:51,016 DEBUG subprocess output: b' Metadata:'
2025-01-14 13:14:51,016 DEBUG subprocess output: b' encoder : Lavc61.19.100 aac'
2025-01-14 13:14:51,016 DEBUG subprocess output: b'[out#0/adts @ 0000021DBE288180] video:0KiB audio:167KiB subtitle:0KiB other streams:0KiB global headers:0KiB muxing overhead: 1.705801%'
2025-01-14 13:14:51,016 DEBUG subprocess output: b'size= 170KiB time=00:00:17.72 bitrate= 78.6kbits/s speed= 176x'
2025-01-14 13:14:51,016 DEBUG subprocess output: b'[aac @ 0000021DBE2B0B80] Qavg: 940.423'
25%|████████████████████████ | 3/12 [00:16<00:52, 5.81s/it]2025-01-14 13:14:51,304 INFO synthesis text 短期拼智力、中期拼毅力、长期拼体力。不要觉得自己落后了，用马拉松的心态，过自己的一生，你慢

新版本windows conda虽然能运行了,但是推理阶段一堆的输出,对比之前的版本是干净的,速度也比之前gradio==4.32.2 的时候慢了10%左右