Our research interests lie in the areas of intelligent speech processing as well as multimodal behavior signal analysis and interpretation.

1. Intelligent speech processing: speaker verification, speaker diarization, paralinguistic state detection, anti-spoofing countermeasure, speech synthesis, voice conversion, keyword spotting, speech separation, spoken language identification, singing and music signal processing, etc.
2. Multimodal behavior signal analysis and interpretation: Gathering, analyzing, modeling and interpreting multimodal human behavior signals (e.g. speech/language/audio/visual/physiological signal analysis and understanding) for assisted diagnose and treatment of autism spectrum disorders, etc.
3. Pathological speech processing: laryngoscopic audio-visual signal processing, electronic laryngeal voice conversion