search

auspicious3000 / autovc

AutoVC: Zero-Shot Voice Style Transfer with Only Autoencoder Loss
https://arxiv.org/abs/1905.05879
MIT License
990 stars 205 forks source link

For training, how many speakers are required? #73

Closed ghost closed 3 years ago

ghost commented 3 years ago

And for each speaker, how many samples, e.g. in seconds in total? Thanks

auspicious3000 commented 3 years ago

About 15 mins per speaker.