How to run TTS on OpenVoice V2? Missing config.json

[mhruiz]

I've been following the Linux installation guide up until the step of installing unidic. The process completed without any errors. However, when I attempted to build a basic TTS feature using the V2 models, I noticed that the BaseSpeakerTTS class inherits from OpenVoiceBaseClass which expects a config_path argument. Yet, there doesn't seem to be a config.json file in the (v2) base_speakers directory.

Given that the guide mentioned:

In April 2024, we released OpenVoice V2, which includes all features in V1 and has:

I assumed that these V2 models could be utilized for the same tasks as V1. However, I'm unsure about how to proceed. Is it possible to use V2 models for TTS, or is this feature exclusive to V1 models?

Any clarification would be greatly appreciated. Thank you!

[aohanhongzhi]

Before

the same as me

besides
there is no demo_part3.ipynb

After

pull the latest code from the repository ,and you can run the code