Can I control the pitch of each phoneme separately?

anonymous-pits / pits

PITS: Variational Pitch Inference for End-to-end Pitch-controllable TTS without External Pitch Predictor

https://anonymous-pits.github.io/pits/

MIT License

274 stars 34 forks source link

Open majo711 opened 1 year ago

majo711 commented 1 year ago

Thanks for your sharing! If possible, how could I control it?

anonymous-pits commented 1 year ago

Yes, it is possible!

However, it requires a lot of manual effort to obtain the duration of each phoneme, check its pitch, and shift it.

We are currently working on automating this process with the reconstructed Yingram from the Yingram decoder