Re:Whisper AI – train on old TV/movies/podcasts for better s...
by brucethemoose2 on Monday June 19, 2023 @04:07PM
(#36395877)
There have actually been some papers on "better than SOTA" TTS speech models with shockingling good inflection, emotion, voice imitation and such.
But the orgs behind them say they are hesitant to release them due to obvious misuse concerns. And I think the unspoken concern is that the datasets are not clean.
But the orgs behind them say they are hesitant to release them due to obvious misuse concerns. And I think the unspoken concern is that the datasets are not clean.