search

facebookresearch / audio2photoreal

Code and dataset for photorealistic Codec Avatars driven from audio
Other
2.66k stars 250 forks source link

Replancement of fairseq #64

Closed SeanChenxy closed 2 months ago

SeanChenxy commented 4 months ago

Hi, thanks for sharing the excellent work. I am trying to run this repo, but fairseq cannot be correctly imported because of the low version of GLIBC. I cannot update my environment. Hene, I wonder whether there are other methods without the use of fairseq. Thanks in advance.

alexanderrichard commented 4 months ago

Hey! You don't need fairseq. You could just copy the model definition from fairseq into a file so you can load the pretrained wav2vec model. Alternatively, if you want to train from scratch, you can plug in any audio encoder of your choice, e.g. a wav2vec from another codebase than fairseq that you might have access to.