search

ZhikangNiu / encodec-pytorch

unofficial implementation of the High Fidelity Neural Audio Compression
MIT License
131 stars 13 forks source link

Additional config details from the hugging face checkpoints #17

Closed darius522 closed 10 months ago

darius522 commented 10 months ago

Hi there,

Thanks a bunch for your effort on that. This is fantastic work.

I was wondering if you could provide a bit more details about the configuration you've used in training the checkpoints you've provided on hugging face ? They sound great and I'd like to re-train them for my own purpose. From their file names, I can infer the following: batch_size=12, tensor_cut=100000, and lr=0.0001, is this right ? What about warmup_epoch, for example ? Additionally, did you use only a subset of LibriTTS or the full 960 hours ?

Thanks again !

ZhikangNiu commented 10 months ago

First, I used full 960h LibriTTS to train the codec model. Second, batch_size=12, tensor_cut=100000, and lr=0.0001 More detailed config information, I will check the node( I wish I hadn't deleted them @darius522