Hey nice one! I'm trying to innovate in the STT and TTS space as well (completely different field), feel free to contact me in my profile email if you want to exchange some knowledge :). I hear some bugs in portuguese, but I'm guessing the trained model has some issues. Congrats and Good luck with your tool! Ps: give users some preview of the speakers voice so we can test it before convert audio, it can save you some resources.
Hello, we can meet on twitter: grsl_en :) thanks for your comment, it's still a v1 in beta but I will continue to improve it.
And yes as you can see in the ui there are buttons for preview but not yet functional :)
