zhangzjn/APB2Face

audio preprocess

eddiecong opened this issue · 0 comments

Hi, I am very interested in your great research progress, I have successfully trained the landmark2face model, but for the audio2landmark model, I have no idea about how to get audio frame MFCC features corresponding to the image frame. Could you share your codes or source codes for preprocessing audio data? Thanks so much and sorry to bother.