model trained on 4 datasets:
ravdess: https://www.kaggle.com/datasets/uwrfkaggler/ravdess-emotional-speech-audio
tess: https://www.kaggle.com/datasets/ejlok1/toronto-emotional-speech-set-tess
savee: https://www.kaggle.com/datasets/ejlok1/surrey-audiovisual-expressed-emotion-savee
crema: https://www.kaggle.com/datasets/ejlok1/cremad
setup:
-
git clone [email protected]:winay007/Speech-emotion-recognition.git.
-
download all 4 datasets provided above.
-
install required packages mentioned in requirements.txt.
-
Run cp-speech-recognition-data-prep.ipynb for creating dataset from audio files.
-
Finally the cp-speech-recognition-model-training.ipynb contains model training code. Go ahead and try different methods for better results or try mine one.