EEG-Based Personalized Emotion Recognition and Receptive Music Therapy System in the Aid of Autism Patients

Autism Spectrum Disorder (ASD) frequently causes emotional regulation problems, which makes it difficult for people to manage and handle emotions such as stress, frustration, and sadness. To overcome these problems, this research provides a therapeutic method that combines music emotion detection with EEG-based emotion recognition. The system is made to identify emotions from EEG(electroencephalogram) signals, connect those emotions to an organized music library, and suggest particular musical selections to help handle emotional states. Our method offers autistic patients a personalized and real-time intervention for emotional dysregulation by using deep learning models, particularly LSTM and CNN architectures on the DEAP (Database for Emotion Analysis using Physiological Signals) and DEAM (Database for Emotional Analysis in Music) datasets. The goal of this paper is to promote emotional stability and also increase mental health through therapeutic and nonverbal encounters by creating a music recommendation system that will counteract the emotions of autistic patients.