The HKCUPU system for the NIST 2010 speaker recognition evaluation

Weiwu Jiang; Man Wai Mak; Wei Rao; Helen Meng

doi:10.1109/ICASSP.2011.5947551

The HKCUPU system for the NIST 2010 speaker recognition evaluation

Weiwu Jiang, Man Wai Mak, Wei Rao, Helen Meng

The Hong Kong Polytechnic University

Research output: Chapter in book / Conference proceeding › Conference article published in proceeding or book › Academic research › peer-review

1 Citation (Scopus)

Abstract

This paper presents the HKCUPU speaker recognition system submitted to NIST 2010 speaker recognition evaluation (SRE). The system comprises five subsystems, each with different acoustic features, session-variability reduction methods, speaker modeling and scoring methods and classifiers. This paper reports the results of individual and fusion systems for the core test and highlights the improvements made by our newly proposed JFA-Fishervoice (FSH) subsystem. Results show that FSH outperforms JFA when its projection matrix is channel-dependent (telephone or microphone) and that FSH is complementary to other state-of-the-art techniques. It was also found that VAD is an important pre-processing step for interview speech.

Original language	English
Title of host publication	2011 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2011 - Proceedings
Pages	5288-5291
Number of pages	4
DOIs	https://doi.org/10.1109/ICASSP.2011.5947551
Publication status	Published - 18 Aug 2011
Event	36th IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2011 - Prague, Czech Republic Duration: 22 May 2011 → 27 May 2011

Conference

Conference	36th IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2011
Country/Territory	Czech Republic
City	Prague
Period	22/05/11 → 27/05/11

Keywords

discriminative models
factor analysis
Fishervoice
NIST SRE 2010
speaker recognition

ASJC Scopus subject areas

Software
Signal Processing
Electrical and Electronic Engineering

More information

10.1109/ICASSP.2011.5947551

Cite this

@inproceedings{47a946eb26a04d3b8e9ff04c175e7d0c,

title = "The HKCUPU system for the NIST 2010 speaker recognition evaluation",

abstract = "This paper presents the HKCUPU speaker recognition system submitted to NIST 2010 speaker recognition evaluation (SRE). The system comprises five subsystems, each with different acoustic features, session-variability reduction methods, speaker modeling and scoring methods and classifiers. This paper reports the results of individual and fusion systems for the core test and highlights the improvements made by our newly proposed JFA-Fishervoice (FSH) subsystem. Results show that FSH outperforms JFA when its projection matrix is channel-dependent (telephone or microphone) and that FSH is complementary to other state-of-the-art techniques. It was also found that VAD is an important pre-processing step for interview speech.",

keywords = "discriminative models, factor analysis, Fishervoice, NIST SRE 2010, speaker recognition",

author = "Weiwu Jiang and Mak, {Man Wai} and Wei Rao and Helen Meng",

year = "2011",

month = aug,

day = "18",

doi = "10.1109/ICASSP.2011.5947551",

language = "English",

isbn = "9781457705397",

pages = "5288--5291",

booktitle = "2011 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2011 - Proceedings",

note = "36th IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2011 ; Conference date: 22-05-2011 Through 27-05-2011",

}

Jiang, W, Mak, MW, Rao, W & Meng, H 2011, The HKCUPU system for the NIST 2010 speaker recognition evaluation. in 2011 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2011 - Proceedings., 5947551, pp. 5288-5291, 36th IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2011, Prague, Czech Republic, 22/05/11. https://doi.org/10.1109/ICASSP.2011.5947551

The HKCUPU system for the NIST 2010 speaker recognition evaluation. / Jiang, Weiwu; Mak, Man Wai; Rao, Wei et al.
2011 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2011 - Proceedings. 2011. p. 5288-5291 5947551.

Research output: Chapter in book / Conference proceeding › Conference article published in proceeding or book › Academic research › peer-review

TY - GEN

T1 - The HKCUPU system for the NIST 2010 speaker recognition evaluation

AU - Jiang, Weiwu

AU - Mak, Man Wai

AU - Rao, Wei

AU - Meng, Helen

PY - 2011/8/18

Y1 - 2011/8/18

N2 - This paper presents the HKCUPU speaker recognition system submitted to NIST 2010 speaker recognition evaluation (SRE). The system comprises five subsystems, each with different acoustic features, session-variability reduction methods, speaker modeling and scoring methods and classifiers. This paper reports the results of individual and fusion systems for the core test and highlights the improvements made by our newly proposed JFA-Fishervoice (FSH) subsystem. Results show that FSH outperforms JFA when its projection matrix is channel-dependent (telephone or microphone) and that FSH is complementary to other state-of-the-art techniques. It was also found that VAD is an important pre-processing step for interview speech.

AB - This paper presents the HKCUPU speaker recognition system submitted to NIST 2010 speaker recognition evaluation (SRE). The system comprises five subsystems, each with different acoustic features, session-variability reduction methods, speaker modeling and scoring methods and classifiers. This paper reports the results of individual and fusion systems for the core test and highlights the improvements made by our newly proposed JFA-Fishervoice (FSH) subsystem. Results show that FSH outperforms JFA when its projection matrix is channel-dependent (telephone or microphone) and that FSH is complementary to other state-of-the-art techniques. It was also found that VAD is an important pre-processing step for interview speech.

KW - discriminative models

KW - factor analysis

KW - Fishervoice

KW - NIST SRE 2010

KW - speaker recognition

UR - http://www.scopus.com/inward/record.url?scp=80051651779&partnerID=8YFLogxK

U2 - 10.1109/ICASSP.2011.5947551

DO - 10.1109/ICASSP.2011.5947551

M3 - Conference article published in proceeding or book

SN - 9781457705397

SP - 5288

EP - 5291

BT - 2011 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2011 - Proceedings

T2 - 36th IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2011

Y2 - 22 May 2011 through 27 May 2011

ER -

The HKCUPU system for the NIST 2010 speaker recognition evaluation

Abstract

Conference

Keywords

ASJC Scopus subject areas

More information

Other files and links

Fingerprint

Cite this