Roformer separation failed: The size of tensor a (60417) must match the size of tensor b (352800) at non-singleton dimension 1

[ColbyDash]

I get this error when trying to use any one of the roformer models, could extend to any of the other models out there, with any clip under 8 seconds long. Oddly enough, the UVR app doesn't have this problem and lets me isolate clips as short as just half a second.

Traceback (most recent call last):
  File "E:\AI\UVR5-UI-1.7.0\app.py", line 283, in roformer_separator
    separation = separator.separate(audio)
  File "E:\AI\UVR5-UI-1.7.0\env\lib\site-packages\audio_separator\separator\separator.py", line 771, in separate
    output_files = self.model_instance.separate(audio_file_path, primary_output_name, secondary_output_name)
  File "E:\AI\UVR5-UI-1.7.0\env\lib\site-packages\audio_separator\separator\architectures\mdxc_separator.py", line 139, in separate
    source = self.demix(mix=mix)
  File "E:\AI\UVR5-UI-1.7.0\env\lib\site-packages\audio_separator\separator\architectures\mdxc_separator.py", line 267, in demix
    result = self.overlap_add(result, x, window, result.shape[-1] - chunk_size, length)
  File "E:\AI\UVR5-UI-1.7.0\env\lib\site-packages\audio_separator\separator\architectures\mdxc_separator.py", line 203, in overlap_add
    result[..., start : start + length] += x[..., :length] * weights[:length]
RuntimeError: The size of tensor a (60417) must match the size of tensor b (352800) at non-singleton dimension 1

The above exception was the direct cause of the following exception:

Traceback (most recent call last):
  File "E:\AI\UVR5-UI-1.7.0\env\lib\site-packages\gradio\queueing.py", line 624, in process_events
    response = await route_utils.call_process_api(
  File "E:\AI\UVR5-UI-1.7.0\env\lib\site-packages\gradio\route_utils.py", line 323, in call_process_api
    output = await app.get_blocks().process_api(
  File "E:\AI\UVR5-UI-1.7.0\env\lib\site-packages\gradio\blocks.py", line 2019, in process_api
    result = await self.call_function(
  File "E:\AI\UVR5-UI-1.7.0\env\lib\site-packages\gradio\blocks.py", line 1566, in call_function
    prediction = await anyio.to_thread.run_sync(  # type: ignore
  File "E:\AI\UVR5-UI-1.7.0\env\lib\site-packages\anyio\to_thread.py", line 56, in run_sync
    return await get_async_backend().run_sync_in_worker_thread(
  File "E:\AI\UVR5-UI-1.7.0\env\lib\site-packages\anyio\_backends\_asyncio.py", line 2505, in run_sync_in_worker_thread
    return await future
  File "E:\AI\UVR5-UI-1.7.0\env\lib\site-packages\anyio\_backends\_asyncio.py", line 1005, in run
    result = context.run(func, *args)
  File "E:\AI\UVR5-UI-1.7.0\env\lib\site-packages\gradio\utils.py", line 865, in wrapper
    response = f(*args, **kwargs)
  File "E:\AI\UVR5-UI-1.7.0\env\lib\site-packages\gradio\utils.py", line 865, in wrapper
    response = f(*args, **kwargs)
  File "E:\AI\UVR5-UI-1.7.0\app.py", line 293, in roformer_separator
    raise RuntimeError(f"Roformer separation failed: {e}") from e
RuntimeError: Roformer separation failed: The size of tensor a (60417) must match the size of tensor b (352800) at non-singleton dimension 1

[Eddycrack864]

Yes, there is currently a problem with Mel/BS Roformers when you use audio with a very short duration, around <10s. It doesn't affect all Roformers but it affects most of them.

But, tbh I don't know how to fix this problem yet, the code is almost identical to the UVR GUI so fixing this may take a while.

[ColbyDash]

Yes, there is currently a problem with Mel/BS Roformers when you use audio with a very short duration, around <10s. It doesn't affect all Roformers but it affects most of them.

But, tbh I don't know how to fix this problem yet, the code is almost identical to the UVR GUI so fixing this may take a while.

I see. In the meantime, I had tried some of the older versions of UVR5 UI to see if they were any different whatsoever. Lo and behold, they weren't but v1.4.0 as far as I know had spat out a bit of a different wording of this error. If it helps in any way, here's that error for you:

Traceback (most recent call last):
  File "E:\AI\UVR5-UI-1.4.0\env\lib\runpy.py", line 196, in _run_module_as_main
    return _run_code(code, main_globals, None,
  File "E:\AI\UVR5-UI-1.4.0\env\lib\runpy.py", line 86, in _run_code
    exec(code, run_globals)
  File "E:\AI\UVR5-UI-1.4.0\env\Scripts\audio-separator.exe\__main__.py", line 7, in <module>
  File "E:\AI\UVR5-UI-1.4.0\env\lib\site-packages\audio_separator\utils\cli.py", line 208, in main
    output_files = separator.separate(audio_file, primary_output_name=args.primary_output_name, secondary_output_name=args.secondary_output_name)
  File "E:\AI\UVR5-UI-1.4.0\env\lib\site-packages\audio_separator\separator\separator.py", line 774, in separate
    output_files = self.model_instance.separate(audio_file_path, primary_output_name, secondary_output_name)
  File "E:\AI\UVR5-UI-1.4.0\env\lib\site-packages\audio_separator\separator\architectures\mdxc_separator.py", line 139, in separate
    source = self.demix(mix=mix)
  File "E:\AI\UVR5-UI-1.4.0\env\lib\site-packages\audio_separator\separator\architectures\mdxc_separator.py", line 267, in demix
    result = self.overlap_add(result, x, window, result.shape[-1] - chunk_size, length)
  File "E:\AI\UVR5-UI-1.4.0\env\lib\site-packages\audio_separator\separator\architectures\mdxc_separator.py", line 203, in overlap_add
    result[..., start : start + length] += x[..., :length] * weights[:length]
RuntimeError: The size of tensor a (60417) must match the size of tensor b (352800) at non-singleton dimension 1
Traceback (most recent call last):
  File "E:\AI\UVR5-UI-1.4.0\env\lib\site-packages\gradio\queueing.py", line 624, in process_events
    response = await route_utils.call_process_api(
  File "E:\AI\UVR5-UI-1.4.0\env\lib\site-packages\gradio\route_utils.py", line 323, in call_process_api
    output = await app.get_blocks().process_api(
  File "E:\AI\UVR5-UI-1.4.0\env\lib\site-packages\gradio\blocks.py", line 2018, in process_api
    result = await self.call_function(
  File "E:\AI\UVR5-UI-1.4.0\env\lib\site-packages\gradio\blocks.py", line 1567, in call_function
    prediction = await anyio.to_thread.run_sync(  # type: ignore
  File "E:\AI\UVR5-UI-1.4.0\env\lib\site-packages\anyio\to_thread.py", line 56, in run_sync
    return await get_async_backend().run_sync_in_worker_thread(
  File "E:\AI\UVR5-UI-1.4.0\env\lib\site-packages\anyio\_backends\_asyncio.py", line 2461, in run_sync_in_worker_thread
    return await future
  File "E:\AI\UVR5-UI-1.4.0\env\lib\site-packages\anyio\_backends\_asyncio.py", line 962, in run
    result = context.run(func, *args)
  File "E:\AI\UVR5-UI-1.4.0\env\lib\site-packages\gradio\utils.py", line 846, in wrapper
    response = f(*args, **kwargs)
  File "E:\AI\UVR5-UI-1.4.0\app.py", line 194, in roformer_separator
    stem1_file = files_list[0]
IndexError: list index out of range