Online feature compensation using modified quantile based noise estimation for robust speech recognition

被引：0

作者：

Lee, Heungkyu ^{[1
]}

Kwon, Ohil ^{[1
]}

Kim, June ^{[1
]}

机构：

[1] Korea Univ, Dept Elect & Comp Engn, Seoul, South Korea

来源：

ADVANCES IN INTELLIGENT IT: ACTIVE MEDIA TECHNOLOGY 2006 | 2006年 / 138卷

关键词：

D O I：

暂无

中图分类号：

TP [自动化技术、计算机技术];

学科分类号：

0812 ;

摘要：

This paper proposes online adaptation method using the modified quantile based noise estimation for feature compensation that is based on Gaussian mixture model for robust speech recognition interface. Proposed method is designed for active online model adaptation method to cope with varying environmental noise conditions. This method is compensated on logarithmic filter-bank energies domain, and modified quantile based noise estimation using beta-order harmonic mean is applied for online noise estimation. Experimental evaluation is done by using Aurora 2 database, and it showed that the proposed method is more robust than other comparative algorithms.

引用

页码：236 / 242

页数：7

共 50 条

[31] On stochastic feature and model compensation approaches to robust speech recognition
Lee, CH
SPEECH COMMUNICATION, 1998, 25 (1-3) : 29 - 47
[32] Two-domain feature compensation for robust speech recognition
Shen, HF
Liu, G
Guo, J
Li, QX
ADVANCES IN NEURAL NETWORKS - ISNN 2005, PT 2, PROCEEDINGS, 2005, 3497 : 351 - 356
[33] Noise robust isolated word recognition using speech feature enhancement techniques
Ecole Nationale d'Ingénieurs de Sfax ENIS, Department of Génie Electrique, BP W, 3038 Sfax, Tunisia
不详
J. Appl. Sci., 2007, 24 (3935-3942):
[34] Modified feature extraction methods in robust speech recognition
Rajnoha, Josef
Pollak, Petr
2007 17TH INTERNATIONAL CONFERENCE RADIOELEKTRONIKA, VOLS 1 AND 2, 2007, : 337 - +
[35] Speech Feature Compensation in Multiple Model Based Speech Recognition System Using VTS-based Environmental Parameter Estimation
Chung, Yongjoo
2013 INTERNATIONAL CONFERENCE ON COMPUTER APPLICATIONS TECHNOLOGY (ICCAT), 2013,
[36] Cepstral statistics compensation and normalization using online pseudo stereo codebooks for robust speech recognition in additive noise environments
Hung, Jeih-Weih
IEICE TRANSACTIONS ON INFORMATION AND SYSTEMS, 2008, E91D (02) : 296 - 311
[37] The Teager Energy based feature parameters for robust speech recognition in car noise
Jabloun, F
Çetin, AE
ICASSP '99: 1999 IEEE INTERNATIONAL CONFERENCE ON ACOUSTICS, SPEECH, AND SIGNAL PROCESSING, PROCEEDINGS VOLS I-VI, 1999, : 273 - 276
[38] Teager energy based feature parameters for robust speech recognition in car noise
Jabloun, Firas
Cetin, A.Enis
ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings, 1999, 1 : 273 - 276
[39] Effective energy feature compensation using modified log-energy dynamic range normalization for robust speech recognition
Lee, Yoonjae
Ko, Hanseok
IEICE TRANSACTIONS ON COMMUNICATIONS, 2007, E90B (06) : 1508 - 1511
[40] Mixed environment compensation based on maximum a posteriori estimation for robust speech recognition
Shen, Haifeng
Liu, Gang
Guo, Jun
ARTIFICIAL INTELLIGENCE REVIEW, 2009, 32 (1-4) : 1 - 11

← 1 2 3 4 5 →