Published numerous papers at top-tier venues including ICASSP, Interspeech, ASRU, SLT, and IEEE Access
Notable works include LLM-based multi-talker ASR, end-to-end neural speaker diarization, audio difference learning for captioning, and non-autoregressive intermediate attractors for diarization
Co-developed the DnR-nonverbal dataset for cinematic audio source separation with non-verbal sounds
Presented research on foley sound synthesis using class-conditioned latent diffusion models at DCASE 2023 Workshop
Co-delivered a tutorial at ICASSP 2021 on distant conversational speech recognition and trends toward end-to-end optimization