TTS models.

[sruckh]

I have trained an F5-TTS model. It seemed possible to use that model with the TTS speech option. I did not see a way to upload a model, and when I searched the directory structure by model name, I could not find the files. So, if we want to use our own model for TTS, how do we do that? It seemed like the RVC models are not used with the TTS functionality.

Thank You.

[JackismyShephard]

@sruckh Currently Ultimate RVC only supports edge TTS. After Edge TTS is applied the resulting voice is converted using RVC. What you can do is:

• apply the F5 TTS model yourself (via CLI or another app).
• go to the multi-step tab under speech conversion in Ultimate RVC
• then go to step 2 and upload your converted speech there

Personally, I am not very impressed with the combination of Edge TTS + RVC , so I think using a more SOTA TTS model might be beneficial. Conceptually it should not be too difficult to implement but requires quite a lot of changes on the frontend side to support multiple models. You are welcome to open a PR if you feel like it. I am not currently working on adding new features to this project.