Dynamic and static models for real-time facial emotion recognition
-
Updated
Aug 2, 2024 - Jupyter Notebook
Dynamic and static models for real-time facial emotion recognition
his is a Speech Emotion Recognition system that classifies emotions from speech samples using deep learning models. The project uses four datasets: CREMAD, RAVDESS, SAVEE, and TESS. The model achieves an accuracy of 96% by combining CNN, LSTM, and CLSTM architectures, along with data augmentation techniques and feature extraction methods.
This project detects depression using audio and visual features from video input. It extracts MFCC features from the full audio and selects 20 evenly spaced frames from the video. These are fused and passed into a DenseNet201 model trained on the DAIC-WOZ dataset. Includes a Gradio web interface, deployable via Hugging Face Spaces and Google Colab.
Add a description, image, and links to the crema-d-dataset topic page so that developers can more easily learn about it.
To associate your repository with the crema-d-dataset topic, visit your repo's landing page and select "manage topics."