
This work aims to establish a robust audio-visual speaker diarization framework that can effectively process content from diverse data domains, addressing challenges such as off-screen speakers, background noise, and domain mismatch.


coremsg

robust-audio-visual-speaker-diarization-addressing-challenges-across-diverse-data-domains


Robust Audio-Visual Speaker Diarization: Addressing Challenges Across Diverse Data Domains


title_rewrite


LoCoNet leverages Long-term Intra-speaker Modeling (LIM) and Short-term Inter-speaker Modeling (SIM) in an interleaved manner to effectively capture both the temporal dependencies of the same speaker and the interactions of speakers in the same scene, achieving state-of-the-art performance on multiple active speaker detection benchmarks.


loconet-an-efficient-long-short-context-network-for-accurate-active-speaker-detection


LoCoNet: An Efficient Long-Short Context Network for Accurate Active Speaker Detection
