OpenBMB
How to TTS Kokoro
1.Hello Is there any document to change the tts I wanted to use the new model Kokoro which is small realtime and accurate. 2. How do the voice clone works as shown in example like talk like trump and its speaks like trump voice
1.Hello Is there any document to change the tts I wanted to use the new model Kokoro which is small realtime and accurate. 2. How do the voice clone works as shown in example like talk like trump and its speaks like trump voice
I think that to use kokoro the wonderful creators of this opensource model would need to retrain the model using kokoro. This is a multimodal model it takes in audio tokens and outputs audio tokens it isn't using intermediary tts models. It is a unified package that AFAIK will not allow you to point to a different tts model. Best wishes and major props to openBMB for the open sourcing such powerful models. We don't deserve them! I for one am super grateful for their contribution.
- Possible but new a lot of training~
- Try to read-> https://github.com/OpenBMB/MiniCPM-o?tab=readme-ov-file#general-speech-conversation-with-configurable-voices
