search

joannahong / Lip2Wav-pytorch

a PyTorch implementation of Lip2Wav
48 stars 11 forks source link

Multi Speaker #5

Closed b21727795 closed 3 years ago

b21727795 commented 3 years ago

Hi! It is great work.

I just have question about Multi Speaker part. Did you finish the implementation of all code. While I trying there is mismatch problem due to model sizes. I took image sizes as 64x64 then this problem solved but another problem pop up. Dimensions not exactly correct I guess.

I just wonder do you work on it or these mistakes are special to me ?

MertCokelek commented 3 years ago

You should update the hparams (embedding input/output dim) to match the data flow. As far as I understood, those parts of the code haven't been completed yet. I'll share the corresponding values soon.

b21727795 commented 3 years ago

Thank you Cokelek