Multimodal data serialization for short term memory

• Main Author: Ananda Theerthan Sripoorani Jayashri
• Other Authors: Justin Dauwels
• Format: Thesis-Master by Coursework
• Language: English
• Published: Nanyang Technological University 2020
• Subjects: Engineering::Mechanical engineering::Robots
• Online Access: https://hdl.handle.net/10356/136987
• Institution: Nanyang Technological University

With the growth of Artificial Intelligence, the applications based on emotion recognition has increased and it has a greater impact in various industries such as Medicine, Psychology and all day to day activities. Interpreting the emotions individually from the conversations, facial expressions, gestures, voice can give different results. In this dissertation, emotion recognition in conversation in integration with the Multi-Modal data is done to increase accuracy and precision. This is implemented in a real-time application by alerting the driver when driving in case he feels overworked and for eLearning sessions for the autism kids to understand their emotions and trying to improve the training given to them. Autistic Children have issues in exposing the emotions in a conversation, which can be dealt with the verbal interjections.