Bug report - [can not use vad]

🐛 Bug

To Reproduce

Steps to reproduce the behavior:

use vad:


(venv) PS D:\Project\autocut> autocut -t --force .\video\1.mp4 --whisper-mode=openai --openai-rpm=50        
[autocut:transcribe.py:L32] INFO   Done Init model in 1.0 sec
[autocut:transcribe.py:L36] INFO   Transcribing .\video\1.mp4
[autocut:utils.py:L115] INFO   .\video\1.md exists. Will overwrite it
Using cache found in C:\Users\myname/.cache\torch\hub\snakers4_silero-vad_master
Traceback (most recent call last):
File "<frozen runpy>", line 198, in _run_module_as_main
File "<frozen runpy>", line 88, in _run_code
File "D:\Project\autocut\venv\Scripts\autocut.exe\__main__.py", line 7, in <module>
File "D:\Project\autocut\venv\Lib\site-packages\autocut\main.py", line 170, in main
Transcribe(args).run()
File "D:\Project\autocut\venv\Lib\site-packages\autocut\transcribe.py", line 42, in run
speech_array_indices = self._detect_voice_activity(audio)
                       ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "D:\Project\autocut\venv\Lib\site-packages\autocut\transcribe.py", line 60, in _detect_voice_activity
self.vad_model, funcs = torch.hub.load(
                        ^^^^^^^^^^^^^^^
File "D:\Project\autocut\venv\Lib\site-packages\torch\hub.py", line 558, in load
model = _load_local(repo_or_dir, model, *args, **kwargs)
        ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "D:\Project\autocut\venv\Lib\site-packages\torch\hub.py", line 587, in _load_local
model = entry(*args, **kwargs)
        ^^^^^^^^^^^^^^^^^^^^^^
File "C:\Users\myname/.cache\torch\hub\snakers4_silero-vad_master\hubconf.py", line 46, in silero_vad
model = init_jit_model(os.path.join(model_dir, 'silero_vad.jit'))
        ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "C:\Users\myname/.cache\torch\hub\snakers4_silero-vad_master\utils_vad.py", line 149, in init_jit_model
model = torch.jit.load(model_path, map_location=device)
        ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "D:\Project\autocut\venv\Lib\site-packages\torch\jit\_serialization.py", line 162, in load
cpp_module = torch._C.import_ir_module(cu, str(f), map_location, _extra_files, _restore_shapes)  # type: ignore[call-arg]
             ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
RuntimeError: open file failed because of errno 42 on fopen: Illegal byte sequence, file path: C:\Users\myname/.cache\torch\hub\snakers4_silero-vad_master\files\silero_vad.jit

2. do not use vad:

autocut -t --force .\video\1.mp4 --whisper-mode=openai --openai-rpm=50 --vad=0


In this case, autocut can run well.

[autocut](https://github.com/mli/autocut)

## Environment

StatusCode        : 200
StatusDescription : OK
Content           : 
                    # Unlike the rest of the PyTorch this file must be python2 compliant.
                    # This script outputs relevant system environment info
                    # Run it with `python collect_env.py`.
                    import datetime
                    import locale
                    impor...
RawContent        : HTTP/1.1 200 OK
                    Connection: keep-alive
                    Content-Security-Policy: default-src 'none'; style-src 'unsafe-inline'; sandbox
                    Strict-Transport-Security: max-age=31536000
                    X-Content-Type-Options: nosniff
                    ...
Forms             : {}
Headers           : {[Connection, keep-alive], [Content-Security-Policy, default-src 'none'; style-src 'unsafe-inline'; sandbox], [Strict-Transport-Security, max-age=31536 
                    000], [X-Content-Type-Options, nosniff]...}
Images            : {}
InputFields       : {}
Links             : {}
ParsedHtml        : mshtml.HTMLDocumentClass
RawContentLength  : 21653

I guess this error might be triggered by Win11 ???

snakers4 / silero-vad

Bug report - [can not use vad] #368

🐛 Bug

To Reproduce