yl4579 / StarGANv2-VC

StarGANv2-VC: A Diverse, Unsupervised, Non-parallel Framework for Natural-Sounding Voice Conversion
MIT License
466 stars 110 forks source link

how to change pitch #40

Closed MMMMichaelzhang closed 2 years ago

MMMMichaelzhang commented 2 years ago

Hi,thanks for this project.It is amazing.When I was thinking about converting between male and female voices, I didn't know how to tune the pitch. Now I'm changing the male voice to a female voice, and the pitch is too high. Can you tell me where to adjust it? Thank you very much.

yl4579 commented 2 years ago

The pitch can be lowered by using a reference with a lower pitch. If you use the mapping network, however, the pitch is uncontrollable.

MMMMichaelzhang commented 2 years ago

Thank you for your reply。Do you mean I should lower the input audio pitch before converting it? I tried using librosa to lower the input audio pitch first, then the audio becomes a little weird. @yl4579

MuruganR96 commented 1 year ago

The pitch can be lowered by using a reference with a lower pitch. If you use the mapping network, however, the pitch is uncontrollable.

When I use the style encoder network, target speaker similarity is decreasing. How to tackle pitch and speaker information both in style encode?