iSchool Capstone

Real-time Speech Emotion Detection

Project tags:

data curation

data science & visualization

software development

Research Award

This project focuses on the development of conversational analytics and AI solutions for, aimed at analyzing real-world speech data for insights. Multiple Machine Learning models were implemented, with Convolutional Neural Networks (CNN) demonstrating the highest accuracy improvement of approximately 67%. The CNN model effectively identifies emotions from audio data clips, enhancing conversational analytics within's product SeaMeet. Extensive experiments and evaluations have been conducted to validate the effectiveness of the CNN-based approach, showcasing its potential for real-time applications such as affective computing, virtual assistants, and emotion-aware human-computer interaction.

Project participants:

Nayan Kaushal


Punya Prakash Shetty


Rakshitha KN


Walker Azam