Abstract:
"Current developments in speech emotion recognition, aided by machine learning and artificial intelligence, promise to revolutionize communication channels, especially in emergencies. Nevertheless, these advancements primarily target widely spoken languages, leaving a significant need for systems that cater to languages like Sinhala.
the research project aims to address the gap in the field of automatic voice emotion recognition (AVER) by introducing a system specifically designed for the Sinhala language. The procedure includes identification and extraction of relevant acoustic features, and the design and implementation of a machine learning model for emotion classification. The system is equipped to recognize critical emotional states such as stress, fear, calmness, and urgency from Sinhala speech. Additionally, it is designed to extract incident-related information from the emotion-infused speech and relay this vital data to the appropriate emergency services, significantly enhancing the effectiveness and efficiency of emergency response in Sri Lanka."
