A Unified Multimodal Framework for Emotion Recognition Using Image, Audio, and Text Modalities

Puntos clave

• This research aims to develop a framework that integrates multiple modalities for effective emotion recognition.
• Employed a unified framework for processing image, audio, and text data.
• Utilized a webcam for capturing live emotional expressions from images.
• Implemented format-independent techniques for recognizing audio sentiment.
• The framework successfully captures and analyzes live images for emotional content.
• Audio sentiment recognition shows consistent results across different formats.