search

myshell-ai / OpenVoice

Instant voice cloning by MIT and MyShell.
https://research.myshell.ai/open-voice
MIT License
29.93k stars 2.95k forks source link

se_extractor.py crashes on line 28: "segments, info = model.transcribe(audio_path, beam_size=5, word_timestamps=True)" #323

Open PK268 opened 1 month ago

PK268 commented 1 month ago

Platform: Windows 11 Python Version: 3.9.0

After following the tutorial here for installing OpenVoice on Windows. I cannot run demo_part3 successfully featuring OpenVoice V2. Narrowing down the issue, I found that the program crashes when trying to transcribe my audio clip in the se_extractor.py line 28. Everything works fine when using V1, granted, only after a couple workarounds. I'd be happy to transcribe the audio myself, is that planned to become a feature?

This is one of my first github issues so let me know if you need more info

dosubot[bot] commented 1 month ago

Hello @PK268! I'm here to help you with any bugs, questions, or contributions you have. Let's work together to solve the issue you're facing.

I wasn't able to find specific information about the 'transcribe' function in OpenVoice V2 or any known issues related to it. However, you might want to check if all dependencies are correctly installed and compatible with your setup. The dependencies for OpenVoice V2 are compatible with Python 3.9.0, so that shouldn't be an issue [1]. If you managed to get V1 working with some workarounds, it might be worth trying similar approaches for V2. Unfortunately, I don't have information on whether manual transcription is planned as a feature.

To continue talking to Dosu, mention @dosu.

Blaizzy commented 1 month ago

Try changing the whisper model from medium to tiny.

It worked for me :)