MrBrian Stasak

My research activities include investigating human study speech-based clinical health applications related to illness and voice disorder screening/monitoring, especially using everyday remote smart device AI technology. My work also includes development of new experimental speech-based protocols and software to help isolate important acoustic, linguistic, and affective feature information which can be used to help detect specific illness (e.g., mental health, neurological, respiratory, voice disorders). I collect large clinically validated patient voice disorder and vocal performing arts datasets, which include audio, nasometer, and lung intensity pressure recordings. My work also entails mentoring cross-discipline University of Sydney students and providing guidance on novel School of Health Sciences Honours Projects. Occasionally, I provide expertise in audio voice forensics using both traditional manual listening and modern state-of-the-art automated AI approaches.

Previously, I worked in the specialised field of audio-speech forensics in the defense sector for 6 years, where I investigated a wide range of 'real world' speech/signal challenges using automated machine learning techniques. I have 3 years previous experience working as a speech language therapist in school settings (K-12).