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

Vishnu Monn Baskaran; Yoong Choon Chang; Jonathan Loo; Koksheik Wong; Ming Tao Gan

doi:10.1109/ICCE-TW.2015.7217016

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

Vishnu Monn Baskaran, Yoong Choon Chang, Jonathan Loo, Koksheik Wong, Ming Tao Gan

Research output: Chapter in Book/Report/Conference proceeding › Conference Paper › Other › peer-review

1 Citation (Scopus)

Abstract

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.

Original language	English
Title of host publication	2015 IEEE International Conference on Consumer Electronics - Taiwan, ICCE-TW 2015
Publisher	IEEE, Institute of Electrical and Electronics Engineers
Pages	492-493
Number of pages	2
ISBN (Electronic)	9781479987443
DOIs	https://doi.org/10.1109/ICCE-TW.2015.7217016
Publication status	Published - 20 Aug 2015
Externally published	Yes
Event	IEEE International Conference on Consumer Electronics-Taiwan 2015 - Taipei, Taiwan Duration: 6 Jun 2015 → 8 Jun 2015 Conference number: 2nd https://ieeexplore.ieee.org/xpl/conhome/7170113/proceeding (Proceedings)

Conference

Conference	IEEE International Conference on Consumer Electronics-Taiwan 2015
Abbreviated title	ICCE-TW 2015
Country/Territory	Taiwan
City	Taipei
Period	6/06/15 → 8/06/15
Internet address	https://ieeexplore.ieee.org/xpl/conhome/7170113/proceeding (Proceedings)

Keywords

Bandwidth
Classification algorithms
Containers
Markov processes
Noise
Prediction algorithms
Transient analysis

Access to Document

10.1109/ICCE-TW.2015.7217016

Cite this

@inproceedings{53d880b69e064e99a33a15ac9e7d59b2,

title = "Dominant speaker detection using discrete Markov chain for multi-user video conferencing",

abstract = "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.",

keywords = "Bandwidth, Classification algorithms, Containers, Markov processes, Noise, Prediction algorithms, Transient analysis",

author = "Baskaran, {Vishnu Monn} and Chang, {Yoong Choon} and Jonathan Loo and Koksheik Wong and Gan, {Ming Tao}",

note = "Publisher Copyright: {\textcopyright} 2015 IEEE. Copyright: Copyright 2017 Elsevier B.V., All rights reserved.; IEEE International Conference on Consumer Electronics-Taiwan 2015, ICCE-TW 2015 ; Conference date: 06-06-2015 Through 08-06-2015",

year = "2015",

month = aug,

day = "20",

doi = "10.1109/ICCE-TW.2015.7217016",

language = "English",

pages = "492--493",

booktitle = "2015 IEEE International Conference on Consumer Electronics - Taiwan, ICCE-TW 2015",

publisher = "IEEE, Institute of Electrical and Electronics Engineers",

address = "United States of America",

url = "https://ieeexplore.ieee.org/xpl/conhome/7170113/proceeding",

}

Baskaran, VM, Chang, YC, Loo, J, Wong, K & Gan, MT 2015, Dominant speaker detection using discrete Markov chain for multi-user video conferencing. in 2015 IEEE International Conference on Consumer Electronics - Taiwan, ICCE-TW 2015. IEEE, Institute of Electrical and Electronics Engineers, pp. 492-493, IEEE International Conference on Consumer Electronics-Taiwan 2015, Taipei, Taiwan, 6/06/15. https://doi.org/10.1109/ICCE-TW.2015.7217016

Dominant speaker detection using discrete Markov chain for multi-user video conferencing. / Baskaran, Vishnu Monn; Chang, Yoong Choon; Loo, Jonathan et al.
2015 IEEE International Conference on Consumer Electronics - Taiwan, ICCE-TW 2015. IEEE, Institute of Electrical and Electronics Engineers, 2015. p. 492-493.

Research output: Chapter in Book/Report/Conference proceeding › Conference Paper › Other › peer-review

TY - GEN

T1 - Dominant speaker detection using discrete Markov chain for multi-user video conferencing

AU - Baskaran, Vishnu Monn

AU - Chang, Yoong Choon

AU - Loo, Jonathan

AU - Wong, Koksheik

AU - Gan, Ming Tao

N1 - Conference code: 2nd

PY - 2015/8/20

Y1 - 2015/8/20

N2 - 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.

AB - 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.

KW - Bandwidth

KW - Classification algorithms

KW - Containers

KW - Markov processes

KW - Noise

KW - Prediction algorithms

KW - Transient analysis

UR - http://www.scopus.com/inward/record.url?scp=84959531834&partnerID=8YFLogxK

U2 - 10.1109/ICCE-TW.2015.7217016

DO - 10.1109/ICCE-TW.2015.7217016

M3 - Conference Paper

AN - SCOPUS:84959531834

SP - 492

EP - 493

BT - 2015 IEEE International Conference on Consumer Electronics - Taiwan, ICCE-TW 2015

PB - IEEE, Institute of Electrical and Electronics Engineers

T2 - IEEE International Conference on Consumer Electronics-Taiwan 2015

Y2 - 6 June 2015 through 8 June 2015

ER -

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

Abstract

Conference

Keywords

Access to Document

Other files and links

Cite this