Liat Kaver
Papers
2
Total Citations
34
H-Index
2
About
Liat Kaver is a researcher advancing the field of audio-visual machine perception, with a primary focus on active speaker detection—a critical component for applications ranging from speaker diarization and video re-targeting to speech enhancement and human-robot interaction. Her most significant contribution is the creation of the AVA-ActiveSpeaker dataset, a large-scale, meticulously labeled audio-visual resource that addresses a long-standing gap in the research community. Prior to this work, the absence of such a comprehensive dataset constrained the development and benchmarking of robust algorithms. By providing this foundational resource, Kaver’s work has directly enabled more accurate and reliable video analysis systems, with her seminal 2020 paper accumulating 19 citations and its 2019 supplementary material receiving 15 citations. Her contributions are particularly notable for bridging the gap between audio and visual modalities, offering a gold-standard benchmark that has become essential for researchers tackling real-world challenges in meeting analysis, human-robot interaction, and beyond. Through her dataset and methodological insights, Kaver has established herself as a key enabler of progress in multi-modal perception.
Research Focus
Key Achievements
Top Papers
- 1Ava Active Speaker: An Audio-Visual Dataset for Active Speaker Detection19 citations · 2020
- 2