This project contains an Emotion Recognition System and it leverages deep learning to classify emotions using audio spectral features like MFCCs, mel spectrograms, chromagrams, and spectral contrast, enabling accurate emotional analysis from speech inputs.

In this project we're finding the emotion of a person by its vocal pattern. To achive this we are training our model with numerous dataset which contain voices of different actors. These are dataset we have:-

CREMA-D ( Crowd-sourced Emotional Mutimodal Actors Dataset ) https://www.kaggle.com/datasets/ejlok1/cremad

RAVDESS ( Ryerson Audio-Visual Database of Emotional Speech and Song ) https://www.kaggle.com/datasets/uwrfkaggler/ravdess-emotional-speech-audio

SAVEE( Surrey Audio-Visual Expressed Emotion ) https://www.kaggle.com/datasets/ejlok1/surrey-audiovisual-expressed-emotion-savee

TESS( Toronto emotional speech set ) https://www.kaggle.com/datasets/ejlok1/toronto-emotional-speech-set-tess

Tensorflow, Pytorch, Sklearn