Deep Learning for Speech Emotion Recognition - Dive into the world of emotional resonance with our cutting-edge CNN-based Speech Emotion Recognition model, achieving an impressive 94% accuracy.
This repository contains code for a Speech Emotion Recognition (SER) system implemented in Python using the Librosa library for audio feature extraction and a Convolutional Neural Network (CNN) model built with Keras for emotion classification. The dataset used for training and testing includes audio recordings from various sources such as the Toronto Emotional Speech Set (TESS), RAVDESS, Surrey Audio-Visual Expressed Emotion (SAVEE), and the Crowd Emotion Manifestation dataset (CREMA-D).
Make sure you have the following libraries installed:
librosanumpymatplotlibtensorflowpandasIPython.displayscikit-learn
You can install them using:
pip install librosa numpy matplotlib tensorflow pandas scikit-learn