Context

Dataset created with the aim to predict females and males emotions based on MFCCs values. With this setup (58 values for each emotion) we were been able to get a good 94% accuracy on the female emotions. Check out the related project at https://www.kaggle.com/cracc97/emotional-speech-recognition Leave an upvote if you like our work :)

Content

The values are been extracted from audios taken from the notorious datasets: RAVDESS, CREMA-D, TESS and SAVEE. Linked respectively:

• https://www.kaggle.com/uwrfkaggler/ravdess-emotional-speech-audio
• https://www.kaggle.com/ejlok1/cremad
• https://www.kaggle.com/ejlok1/toronto-emotional-speech-set-tess
• https://www.kaggle.com/ejlok1/surrey-audiovisual-expressed-emotion-savee

Acknowledgements

This project and dataset has been carried out as university exam, so feel free to use this material if you found it useful