HMM Compensation Based on Non-uniform Spectral Compression for Noisy Speech Recognition

被引：0

作者：

Ning, Geng-xin ^{[1
]}

Zhang, Jun ^{[1
]}

Yu, Hua ^{[1
]}

机构：

[1] S China Univ Technol, Sch Elect & Informat Engn, Guangzhou 510640, Peoples R China

来源：

2008 11TH IEEE SINGAPORE INTERNATIONAL CONFERENCE ON COMMUNICATION SYSTEMS (ICCS), VOLS 1-3 | 2008年

关键词：

Model-based compensation; non-uniform spectral compression; noisy speech recognition;

D O I：

10.1109/ICCS.2008.4737168

中图分类号：

TP3 [计算技术、计算机技术];

学科分类号：

0812 ;

摘要：

A robust speech feature extraction method based on the power law of hearing and non-uniform spectral compression technique is proposed, and the correspondent model compensation algorithm is given. The mismatch functions, reflecting the infections of additive noise and spectral compression, and the model compensation formulae are deduced. The experiment results show that the significant improvement is obtained over the popular VTS (Vector Taylor Series) by adopting those approaches. It's concluded that the proposed method can deal with the speech recognition tasks in different additive noisy environments.

引用

页码：184 / 187

页数：4

共 50 条

[21] Accurate compensation in the log-spectral domain for noisy speech recognition
Afify, M
IEEE TRANSACTIONS ON SPEECH AND AUDIO PROCESSING, 2005, 13 (03): : 388 - 398
[22] Non-uniform error criteria for automatic pattern and speech recognition
Fu, Qiang
Mansjur, Dwi Sianto
Juang, Biing-Hwang
2008 IEEE INTERNATIONAL CONFERENCE ON ACOUSTICS, SPEECH AND SIGNAL PROCESSING, VOLS 1-12, 2008, : 1853 - 1856
[23] HCRF-based Model Compensation for Noisy Speech Recognition
Hong, Wei-Tyng
2013 IEEE 17TH INTERNATIONAL SYMPOSIUM ON CONSUMER ELECTRONICS (ISCE), 2013, : 277 - 278
[24] Optimal clustering and non-uniform allocation of Gaussian Kernels in scalar dimension for HMM compression
Li, XB
Soong, FK
Myrvoll, TA
Wang, RH
2005 IEEE INTERNATIONAL CONFERENCE ON ACOUSTICS, SPEECH, AND SIGNAL PROCESSING, VOLS 1-5: SPEECH PROCESSING, 2005, : 669 - 672
[25] AN HMM COMPENSATION APPROACH FOR DYNAMIC FEATURES USING UNSCENTED TRANSFORMATION AND ITS APPLICATION TO NOISY SPEECH RECOGNITION
Hu, Yu
Huo, Qiang
2008 6TH INTERNATIONAL SYMPOSIUM ON CHINESE SPOKEN LANGUAGE PROCESSING, PROCEEDINGS, 2008, : 109 - 112
[26] Automatic generation of non-uniform HMM topologies based on the MDL criterion
Jitsuhiro, T
Matsui, T
Nakamura, S
IEICE TRANSACTIONS ON INFORMATION AND SYSTEMS, 2004, E87D (08): : 2121 - 2129
[27] On combining frequency warping and spectral shaping in HMM based speech recognition
Potamianos, A
Rose, RC
1997 IEEE INTERNATIONAL CONFERENCE ON ACOUSTICS, SPEECH, AND SIGNAL PROCESSING, VOLS I - V: VOL I: PLENARY, EXPERT SUMMARIES, SPECIAL, AUDIO, UNDERWATER ACOUSTICS, VLSI; VOL II: SPEECH PROCESSING; VOL III: SPEECH PROCESSING, DIGITAL SIGNAL PROCESSING; VOL IV: MULTIDIMENSIONAL SIGNAL PROCESSING, NEURAL NETWORKS - VOL V: STATISTICAL SIGNAL AND ARRAY PROCESSING, APPLICATIONS, 1997, : 1275 - 1278
[28] A Trajectory Compression Algorithm Based on Non-uniform Quantization
Lv, Chengjiao
Chen, Feng
Xu, Yongzhi
Song, Junping
Lv, Pin
2015 12TH INTERNATIONAL CONFERENCE ON FUZZY SYSTEMS AND KNOWLEDGE DISCOVERY (FSKD), 2015, : 2469 - 2474
[29] Importance of Non-Uniform Prosody Modification for Speech Recognition in Emotion Conditions
Raju, V. V. Vidyadhara
Vydana, Hari Krishna
Gangashetty, Suryakanth, V
Vuppala, Anil Kumar
2017 ASIA-PACIFIC SIGNAL AND INFORMATION PROCESSING ASSOCIATION ANNUAL SUMMIT AND CONFERENCE (APSIPA ASC 2017), 2017, : 573 - 576
[30] A frequency-weighted HMM based on minimum error classification for noisy speech recognition
Matsumoto, H
Ono, M
1997 IEEE INTERNATIONAL CONFERENCE ON ACOUSTICS, SPEECH, AND SIGNAL PROCESSING, VOLS I - V: VOL I: PLENARY, EXPERT SUMMARIES, SPECIAL, AUDIO, UNDERWATER ACOUSTICS, VLSI; VOL II: SPEECH PROCESSING; VOL III: SPEECH PROCESSING, DIGITAL SIGNAL PROCESSING; VOL IV: MULTIDIMENSIONAL SIGNAL PROCESSING, NEURAL NETWORKS - VOL V: STATISTICAL SIGNAL AND ARRAY PROCESSING, APPLICATIONS, 1997, : 1511 - 1514

← 1 2 3 4 5 →