How to use the Yin algorithm compute GPE FFE

NVIDIA / mellotron

Mellotron: a multispeaker voice synthesis model based on Tacotron 2 GST that can make a voice emote and sing without emotive or singing training data

BSD 3-Clause "New" or "Revised" License

853 stars 184 forks source link

How to use the Yin algorithm compute GPE FFE #11

Closed xiaolkai closed 4 years ago

xiaolkai commented 4 years ago

I want to eval my synthesis audio. The paper said that pitch and voicing metrics are computed using the Yin algorithm. I read the code but i didn't see the evaluation part. Does anyone can share the code to compute GPE and FFE. Thank you!

rafaelvalle commented 4 years ago

Search github for Gross Pitch Error. For example, this repo has an implementation.

rafaelvalle commented 4 years ago

Closing due to inactivity.