A Unified Multimodal Framework for Emotion Recognition Using Image, Audio, and Text Modalities | Synapse
March 28, 2026Open Access
A Unified Multimodal Framework for Emotion Recognition Using Image, Audio, and Text Modalities
Key Points
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.
Abstract
A Unified Multimodal Framework for Emotion recognition Using Image, Audio, and Text modalities. Use of Web cam for live image sentiment capture. Format independent Audio Sentiment recognition.