search

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

Yin pitch set to minimum 100 Hz #19

Closed drfeinberg closed 4 years ago

drfeinberg commented 4 years ago

Your Yin Pitch algorithm pitch minimum is set to 100 Hz, so you can't use any voices lower than that threshold. To lower it, you'll also need to make the window size larger to accommodate the lower frequencies.

rafaelvalle commented 4 years ago

https://github.com/NVIDIA/mellotron/blob/master/hparams.py#L44

drfeinberg commented 4 years ago

Sorry I missed that. Thanks.