search

zdavidli / chirp

Real-time tweet reading with custom-trained voice models
3 stars 0 forks source link

Make syllables proportional in output length #29

Closed zdavidli closed 6 years ago

zdavidli commented 6 years ago

Currently it sounds mechanical because all the phonemes take the same amount of time,

zdavidli commented 6 years ago

Voice model was changed, this issue was irrelevant.