E-Talk: Accelerating Active Speaker Detection with Audio-Visual Fusion and Edge-Cloud Computing
Published in 2023 20th Annual IEEE International Conference on Sensing, Communication, and Networking (SECON), 2023
leverages voiceprint consistency and facial analysis to enhance active speaker identification in surveillance systems.
Recommended citation: Xiaojing Yu, Lan Zhang, and Xiang-Yang Li. E-Talk: Accelerating Active Speaker Detection with Audio-Visual Fusion and Edge-Cloud Computing (SECON 2023).