Jiaxin-Ye / TIM-Net_SER

[ICASSP 2023] Official Tensorflow implementation of "Temporal Modeling Matters: A Novel Temporal Emotional Modeling Approach for Speech Emotion Recognition".
GNU General Public License v3.0
161 stars 25 forks source link

Questions about training #23

Open AntyRia opened 5 months ago

AntyRia commented 5 months ago

Hi, thank you for your great contribution, this project is really great, but I have a few questions about the use process.

I want to train a weight using my own audio data set (which customizes the sentiment category I'm interested in), what should I do, my idea is to define my own label in the 'extract_feature.py' script, and then extract the features, Then use 'main.py' to train directly in 'train' mode. Is this solution feasible? Or is there a better way to do it?

Looking forward to your reply!