[Discussion] Add others voice models

[Brodvd]

Hey @RyanMetcalfeInt8 what do you think about adding text-to-speech and voice-cloning model to OpenVINO? I found Coqui.IA, It seems the best, you can try It also here.
Instead, maybe you have already seen It, It was just published here the pre-trained model MelBand Roformer Vocals DeReverb/DeEcho that, seeing the examples, seems very good as voice-dereverb.

[RyanMetcalfeInt8]

Hi @Brodvd,

Thanks for the tips! I would love to eventually add text-to-speech / voice-cloning models to this project -- It's something that I've wanted to do all along but haven't yet found the time to do.. perhaps next year.

Instead, maybe you have already seen It, It was just published here the pre-trained model MelBand Roformer Vocals DeReverb/DeEcho that, seeing the examples, seems very good as voice-dereverb.

Yes, I absolutely have seen that -- and am currently in the process of converting / optimizing it for use in this project. I think DeReverb is such a useful feature that this is top of my priority list. Stay tuned.

[MarcoRavich]

Just discovered ClearerVoice-Studio by @alibabasglab & @zexupan a toolkit for speech enhancement, speech separation, speech super-resolution, target speaker extraction, and more.

As always you can also find more resources listed in the - currently updating - AI-Voicing page.

Hope that inspires !