Getting "FileNotFoundError: [WinError 3] The system cannot find the path specified: ''" when trying to clone a voice

[bimdas]

Getting this error when trying to clone a voice.

Traceback (most recent call last): File "F:\AI\Bark\installer_files\env\lib\site-packages\gradio\routes.py", line 439, in run_predict output = await app.get_blocks().process_api( File "F:\AI\Bark\installer_files\env\lib\site-packages\gradio\blocks.py", line 1384, in process_api result = await self.call_function( File "F:\AI\Bark\installer_files\env\lib\site-packages\gradio\blocks.py", line 1089, in call_function prediction = await anyio.to_thread.run_sync( File "F:\AI\Bark\installer_files\env\lib\site-packages\anyio\to_thread.py", line 33, in run_sync return await get_asynclib().run_sync_in_worker_thread( File "F:\AI\Bark\installer_files\env\lib\site-packages\anyio_backends_asyncio.py", line 877, in run_sync_in_worker_thread return await future File "F:\AI\Bark\installer_files\env\lib\site-packages\anyio_backends_asyncio.py", line 807, in run result = context.run(func, args) File "F:\AI\Bark\installer_files\env\lib\site-packages\gradio\utils.py", line 700, in wrapper response = f(args, **kwargs) File "F:\AI\Bark\bark\bark_webui.py", line 317, in clone_voice_gradio clone_dir = clone_voice( File "F:\AI\Bark\bark\bark_infinity\clonevoice.py", line 324, in clone_voice audio_filepath_files = os.listdir(audio_filepath_directory) FileNotFoundError: [WinError 3] The system cannot find the path specified: ''

[JonathanFly]

It's a bug, I'll try and fix Saturday night. In the meantime if you go the extras tap and make sure none of the path fields are blank, it should work for now.

[jurandfantom]

would like to report that same error goes for text2voice feature as well - will wait till next week for updates and test again. Wish to be more useful, but its fresh install step by step. Happy to find your project - love GUI and dedication :) wish you best of luck with it.

segment_text: hello world, my name is dog --Segment 1/1: est. 2.40s (1 of 1 iterations) hello world, my name is dog ----> negative_text_prompt: specific_npz_file_negative_prompt: None negative_text_prompt_logits_scale: None negative_text_prompt_divergence_scale: None -->All Audio Sections Finished at: 2023-07-11 17:28:47 in 29.45 seconds Traceback (most recent call last): File "E:\Magazyn\Grafika\AI\Text2Voice\Bark_Suno_v2\installer_files\env\lib\site-packages\gradio\routes.py", line 439, in run_predict output = await app.get_blocks().process_api( File "E:\Magazyn\Grafika\AI\Text2Voice\Bark_Suno_v2\installer_files\env\lib\site-packages\gradio\blocks.py", line 1384, in process_api result = await self.call_function( File "E:\Magazyn\Grafika\AI\Text2Voice\Bark_Suno_v2\installer_files\env\lib\site-packages\gradio\blocks.py", line 1089, in call_function prediction = await anyio.to_thread.run_sync( File "E:\Magazyn\Grafika\AI\Text2Voice\Bark_Suno_v2\installer_files\env\lib\site-packages\anyio\to_thread.py", line 33, in run_sync return await get_asynclib().run_sync_in_worker_thread( File "E:\Magazyn\Grafika\AI\Text2Voice\Bark_Suno_v2\installer_files\env\lib\site-packages\anyio\_backends\_asyncio.py", line 877, in run_sync_in_worker_thread return await future File "E:\Magazyn\Grafika\AI\Text2Voice\Bark_Suno_v2\installer_files\env\lib\site-packages\anyio\_backends\_asyncio.py", line 807, in run result = context.run(func, *args) File "E:\Magazyn\Grafika\AI\Text2Voice\Bark_Suno_v2\installer_files\env\lib\site-packages\gradio\utils.py", line 700, in wrapper response = f(*args, **kwargs) File "E:\Magazyn\Grafika\AI\Text2Voice\Bark_Suno_v2\installer_files\env\lib\site-packages\gradio\utils.py", line 700, in wrapper response = f(*args, **kwargs) File "E:\Magazyn\Grafika\AI\Text2Voice\Bark_Suno_v2\bark\bark_webui.py", line 749, in generate_audio_long_gradio ) = api.generate_audio_long_from_gradio(**kwargs) File "E:\Magazyn\Grafika\AI\Text2Voice\Bark_Suno_v2\bark\bark_infinity\api.py", line 1301, in generate_audio_long_from_gradio ) = generate_audio_long(**kwargs) File "E:\Magazyn\Grafika\AI\Text2Voice\Bark_Suno_v2\bark\bark_infinity\api.py", line 1616, in generate_audio_long write_one_segment( File "E:\Magazyn\Grafika\AI\Text2Voice\Bark_Suno_v2\bark\bark_infinity\api.py", line 505, in write_one_segment write_seg_wav(filepath, audio_arr, **kwargs) File "E:\Magazyn\Grafika\AI\Text2Voice\Bark_Suno_v2\bark\bark_infinity\api.py", line 562, in write_seg_wav write_audiofile(filepath, audio_arr, **kwargs) File "E:\Magazyn\Grafika\AI\Text2Voice\Bark_Suno_v2\bark\bark_infinity\api.py", line 677, in write_audiofile audio = AudioSegment.from_wav(temp_wav) File "E:\Magazyn\Grafika\AI\Text2Voice\Bark_Suno_v2\installer_files\env\lib\site-packages\pydub\audio_segment.py", line 808, in from_wav return cls.from_file(file, 'wav', parameters=parameters) File "E:\Magazyn\Grafika\AI\Text2Voice\Bark_Suno_v2\installer_files\env\lib\site-packages\pydub\audio_segment.py", line 728, in from_file info = mediainfo_json(orig_file, read_ahead_limit=read_ahead_limit) File "E:\Magazyn\Grafika\AI\Text2Voice\Bark_Suno_v2\installer_files\env\lib\site-packages\pydub\utils.py", line 274, in mediainfo_json res = Popen(command, stdin=stdin_parameter, stdout=PIPE, stderr=PIPE) File "E:\Magazyn\Grafika\AI\Text2Voice\Bark_Suno_v2\installer_files\env\lib\subprocess.py", line 971, in __init__ self._execute_child(args, executable, preexec_fn, close_fds, File "E:\Magazyn\Grafika\AI\Text2Voice\Bark_Suno_v2\installer_files\env\lib\subprocess.py", line 1456, in _execute_child hp, ht, pid, tid = _winapi.CreateProcess(executable, args, FileNotFoundError: [WinError 2] The system cannot find the file specified

[JonathanFly]

Just pushed a fix. If you know how to do a git pull you can do so.

If using the windows installer, you should be able just run TROUBLESHOOT_BARK_INFINITY.bat and then do the git pull (make sure you are in the bark directory)

[neonarc4]

@JonathanFly is clone voice not work for directml?

if get_SUNO_USE_DIRECTML() is True:
  ...
else:
   ...
    from .clonevoice import wav_to_semantics, generate_fine_from_wav, quick_clone

since u only define on else statement

[JonathanFly]

@JonathanFly is clone voice not work for directml?

if get_SUNO_USE_DIRECTML() is True:
  ...
else:
   ...
    from .clonevoice import wav_to_semantics, generate_fine_from_wav, quick_clone

since u only define on else statement

It didn't work for directml so I just did that to get Bark overall working. It might not be a hard fix, didn't try and figure it out.

I was kind of hoping DirectML would get a newer version and fix more bugs so it didn't require as many code changes. But I just checked, and it looks like torch-directml==0.2.0.dev230426 is still the latest version? I'm a little confused how to use the latest version on Github honestly. (I only used DirectML to try and get Bark to work this one time, not otherwise familiar with it.)

[neonarc4]

honestly look like these directml people didnt want to go recent torch version 2.1 and torch audio 2.1 but i heard hip sdk launched on rocm no idea about that part ;/ but waiting for rocm torch would be look like an month :_ guess let wait for directml v2.1

here hop in https://github.com/microsoft/DirectML/issues/537