search

TimoBolkart / voca

This codebase demonstrates how to synthesize realistic 3D character animations given an arbitrary speech signal and a static character mesh.
https://voca.is.tue.mpg.de/en
1.14k stars 271 forks source link

How can I make npy and pkl training data for my own dataset? #82

Open Enin opened 2 years ago

Enin commented 2 years ago

I am ready my own wav file and sequential obj mesh datasets(5023x3). voca training session used datasets as .npy and .pkl file and I want to know about how to make those files. Could you tell me how to make? thank you

monokuma-zhuo commented 2 years ago

Do you have any leads? I have the same question.

zhanchao019 commented 2 years ago

I am ready my own wav file and sequential obj mesh datasets(5023x3). voca training session used datasets as .npy and .pkl file and I want to know about how to make those files. Could you tell me how to make? thank you

sorry to bother, but where could i find the raw_audio_fixed.pkl and subj_seq_to_idx.pkl

yangdaowu commented 2 years ago

I have the same question.

wzx7084 commented 1 year ago

I have the same queation.

wangxuanx commented 1 year ago

I also have this question and want to train my dataset. May I ask how to generate a pkl file for WAV voice? May I ask if you can help us answer? @Enin @zhanchao019 @monokuma-zhuo @yangdaowu @wzx7084