Design SER system with the ability to process and classify speech signals to detect emotions embedded

ORI Grant ID
CCB21-2021-00021-003
Research Area(s): Systems Integration | Funder: NSERC | Program: ARTP-1
PI Name: Parisa Pouladzadeh | Faculty/Department: Faculty of Applied Sciences & Technology

Press’nXPress plans to add contextual sentiment analysis to determine what emotion is being expressed, such as customer frustration, confusion, and concern in a phone call conversation. Although there are many advancements in speech emotion recognition (SER) systems, SER performance drastically decreases in natural noisy environments and call center phone quality. In this project, we are going to design and optimize an SER system with the ability to process and classify speech signals to detect emotions embedded in them in a near real-time situation with the call center phone call quality. Design SER system with the ability to process and classify speech signals to detect emotions embedded.