train.py missing & custom data training

JusperLee / CTCNet

An Audio-Visual Speech Separation Model Inspired by Cortico-Thalamo-Cortical Circuits

Apache License 2.0

68 stars 16 forks source link

train.py missing & custom data training #7

Closed AhmadHakami closed 7 months ago

AhmadHakami commented 7 months ago

Hi @JusperLee, thank you for your amazing work!

after taking a look at the README.md and the files inside this repository i could not find the train.py file and im wondering if it is possible to train the model only on an audio data (mixed & separated) without videos

JusperLee commented 7 months ago

I have processed the visual information into npz format and saved it to the mouths folder.