Cepstral domain segmental nonlinear feature transformations for robust speech recognition

被引：38
|
作者：
Segura, JC [1 ]
Benítez, C [1 ]
de la Torre, A [1 ]
Rubio, AJ [1 ]
Ramírez, J [1 ]
机构：
[1] Univ Granada, Dept Elect & Tecnol Comp, E-18071 Granada, Spain
来源：
IEEE SIGNAL PROCESSING LETTERS | 2004年 / 11卷 / 05期
关键词：
histogram equalization; order statistics; robustness; speech recognition;
D O I：
10.1109/LSP.2004.826648
中图分类号：
TM [电工技术]; TN [电子技术、通信技术];
学科分类号：
0808 ; 0809 ;
摘要：
This letter presents a new segmental nonlinear feature normalization algorithm to improve the robustness of speech recognition systems against variations of the acoustic environment. An experimental study of the best delay-performance tradeoff is conducted within the AURORA-2 framework, and a comparison with two commonly used normalization algorithms is presented. Computational IN: efficient algorithms based on order statistics are also presented. One of them is based on linear interpolation between sampling quantiles, and the other one is based on a point estimation of the probability distribution. The reduction in the computational cost does not degrade the performance significantly.
引用
收藏
页码：517 / 520
页数：4
相关论文
共 50 条
  • [1] Cepstral domain segmental feature vector normalization for noise robust speech recognition
    Viikki, O
    Laurila, K
    [J]. SPEECH COMMUNICATION, 1998, 25 (1-3) : 133 - 147
  • [2] Multichannel Cepstral Domain Feature Warping for Robust Speech Recognition
    Squartini, Stefano
    Fagiani, Marco
    Principi, Emanuele
    Piazza, Francesco
    [J]. NEURAL NETS WIRN10, 2011, 226 : 284 - 292
  • [3] CEPSTRAL DOMAIN TALKER STRESS COMPENSATION FOR ROBUST SPEECH RECOGNITION
    CHEN, YN
    [J]. IEEE TRANSACTIONS ON ACOUSTICS SPEECH AND SIGNAL PROCESSING, 1988, 36 (04): : 433 - 439
  • [4] MEDIUM-DURATION MODULATION CEPSTRAL FEATURE FOR ROBUST SPEECH RECOGNITION
    Mitra, Vikramjit
    Franco, Horacio
    Graciarena, Martin
    Vergyri, Dimitra
    [J]. 2014 IEEE INTERNATIONAL CONFERENCE ON ACOUSTICS, SPEECH AND SIGNAL PROCESSING (ICASSP), 2014,
  • [5] Nonlinear spectral transformations for robust speech recognition
    Ikbal, S
    Hermansky, H
    Bourlard, H
    [J]. ASRU'03: 2003 IEEE WORKSHOP ON AUTOMATIC SPEECH RECOGNITION AND UNDERSTANDING ASRU '03, 2003, : 393 - 398
  • [6] Feature Transformations for Robust Speech Recognition in Reverberant Conditions
    Yuliani, Asri R.
    Sustika, Rika
    Yuwana, Raden S.
    Pardede, Hilman F.
    [J]. 2017 INTERNATIONAL CONFERENCE ON COMPUTER, CONTROL, INFORMATICS AND ITS APPLICATIONS (IC3INA), 2017, : 57 - 62
  • [7] FEATURE EXTRACTION ALGORITHM USING NEW CEPSTRAL TECHNIQUES FOR ROBUST SPEECH RECOGNITION
    Korba, Mohamed Cherif Amara
    Bourouba, Houcine
    Djemili, Rafik
    [J]. MALAYSIAN JOURNAL OF COMPUTER SCIENCE, 2020, 33 (02) : 90 - 101
  • [8] Parametric nonlinear feature equalization for robust speech recognition
    Garcia, Luz
    Segura, Jose C.
    Ramirez, Javier
    de la Torre, Angel
    Benitez, Carmen
    [J]. 2006 IEEE INTERNATIONAL CONFERENCE ON ACOUSTICS, SPEECH AND SIGNAL PROCESSING, VOLS 1-13, 2006, : 529 - 532
  • [9] Cepstral Distance and Log Energy Based Silence Feature Normalization for Robust Speech Recognition
    Shen, Guanghu
    Chung, Hyun-Yeol
    [J]. JOURNAL OF THE ACOUSTICAL SOCIETY OF KOREA, 2010, 29 (04): : 278 - 285
  • [10] Multi-Input Feature Combination in the Cepstral Domain for Practical Speech Recognition Systems
    Obuchi, Yasunari
    Hataoka, Nobuo
    [J]. IEICE TRANSACTIONS ON INFORMATION AND SYSTEMS, 2009, E92D (04): : 662 - 670
12345