Dominant speaker detection using discrete Markov chain for multi-user video conferencing

Baskaran, Vishnu Monn, Choon Chang, Yoong, Loo, Jonathan ORCID:, Wong, KokSheik and Gan, Ming-Tao (2015) Dominant speaker detection using discrete Markov chain for multi-user video conferencing. In: 2015 IEEE International Conference on Consumer Electronics - Taiwan (ICCE-TW), 06-08 June 2015, Taipei, Taiwan.

[thumbnail of Baskaran-etal-2015-Dominant-speaker-detection-using-discrete-Markov-chain.pdf] PDF
Baskaran-etal-2015-Dominant-speaker-detection-using-discrete-Markov-chain.pdf - Published Version
Restricted to Repository staff only

Download (255kB)


This paper puts forward a discrete-time Markov chain algorithm in predicting a pair of active or dominant speakers in an ultra-high definition multi-user video conferencing system. The applied Markov chain minimizes false dominant speaker classification due to transient noise during a video conferencing session. This algorithm also includes a set of linear weights-based assignment for both the initial state vector and transition probability matrix, which improves the response of the algorithm towards changing dominant speakers. Experimental results suggests that this algorithm accurately predicts the most dominant speaker at a rate of 83% for 11 clients in a combined video with 86% reduction in false dominant speaker classification, based on given a set of artificial speaker data.

Item Type: Conference or Workshop Item (Paper)
ISBN: 9781479987450
Identifier: 10.1109/ICCE-TW.2015.7217016
Page Range: pp. 492-493
Identifier: 10.1109/ICCE-TW.2015.7217016
Keywords: Prediction algorithms, Classification algorithms, Markov processes, Bandwidth, Transient analysis, Noise, Containers
Subjects: Computing > Systems
Depositing User: Jonathan Loo
Date Deposited: 21 Jun 2017 15:35
Last Modified: 28 Aug 2021 07:23

Actions (login required)

View Item View Item