Emotion Detection Application

This emotion detection application uses computer vision and speech processing techniques, leveraging TensorFlow and Keras. For face emotion detection, it classifies facial expressions in grayscale images into seven categories using a convolutional neural network (CNN). The model is trained with data augmentation and a data generator for efficient processing. For audio emotion detection, it processes audio features such as RMS, zero crossing rate, and MFCCs, feeding them into a CNN to classify emotions from speech into eight categories. Both models are evaluated using metrics like accuracy, loss, and confusion matrices, enhancing the application's robustness in recognizing emotions from visual and audio inputs to support various user interactions.

Check it out