Comment on Deep Voice: Real-Time Neural Text-To-SpeechComments−whodunser9yIt says they trained on 20 hours of a speech corpus subset. Will larger datasets influence the future of TTS?
Comments
It says they trained on 20 hours of a speech corpus subset. Will larger datasets influence the future of TTS?