[PDF]

An Android-Based Deep Learning System for Emotion Recognition: Face Detection and Speech Sentiment Fusion


Wenzhi Hu

08/05/2025

Supervised by Hiroyuki Kido; Moderated by Mohammad Taher Pilehvar

This project aims to develop an Android-based emotion recognition application that integrates facial detection in camera feeds (using Resnet-18 or other object detection algorithms) with speech sentiment emotion analysis (MFCC+MLP). By fusing the outputs, the app will provide a comprehensive result to the user's current emotional state and generate results to the users.


Initial Plan (03/02/2025) [Zip Archive]

Final Report (08/05/2025) [Zip Archive]

Publication Form