Synthesized Speech Detection Under Emotional Voice Conditions
The project aims on the development of a countermeasure system that can handle emotional speech input.
Project GitHub: Not Ready.
The project is conducted by my 3rd year student, Zykova-Myzina Anna, at HSE (Coursework) in 24/25 academic year.
Countermeasure (CM) systems distinguish between real and synthesized speech. However, existing solutions are mostly trained on datasets with neutral voice. This means that emotional speech, such as crying or screaming, will be out-of-domain for these systems and lead to poor performance. The project goal is to explore and improve the quality of CMs under emotional voice conditions.