search

CorentinJ / Real-Time-Voice-Cloning

Clone a voice in 5 seconds to generate arbitrary speech in real-time
Other
52.18k stars 8.73k forks source link

I keep getting TypeError: Invalid file: WindowsPath #235

Closed Delo001 closed 4 years ago

Delo001 commented 4 years ago

Hi guys I keep getting this error when running python demo_toolbox.py Exception: Invalid file: WindowsPath('D:/ai/LibriSpeech/train-clean-360/6157/40556/6157-40556-0111.flac')

Also I get this error using eg. python synthesizer_preprocess_audio.py TypeError: Invalid file: WindowsPath('D:/ai/LibriSpeech/train-clean-100/103/1240/103-1240-0000.flac')

Any help with this would be fantastic it may be something simple I have only just started with python a few days ago.

I am running Windows 10 and using Anaconda and have downloaded all the files required. I just cant seem to load any voices in the toolbox through Voxceleb, librispeech or custom audio files in any format but i can record my own voice in toolbox. Thanks guys hopefully someone can help me out.

Cheers Glenn

SpiderKnot commented 4 years ago

Hi! Have you solved this issue? I'm facing the exact the same problem.

Kayzh3r commented 4 years ago

any news about this issue?

SamuelDietz commented 4 years ago

Bump. This issue is affecting me aswell!

Kayzh3r commented 4 years ago

This bug is easy to debug. I was testing the algorithm so I hard-coded the url of the input files with the raw string modifier in python. E.g., r"PATH2MYFILE"

SamuelDietz commented 4 years ago

Actually I just fixed it for myself.

I changed line 119 in toolbox/init.py to: wav = Synthesizer.load_preprocess_wav(str(fpath)) now it works without any problems.

ghost commented 4 years ago

It's nice to see you all found a workaround. We formally resolved this problem in #371 . Closing this issue.