Correction of likelihoods for degrees of freedom in robust speech recognition using missing feature theory

被引：0
|
作者：
Van hamme, H [1 ]
机构：
[1] Katholieke Univ Leuven, Dept ESAT, B-3001 Heverlee, Belgium
来源：
SEVENTH INTERNATIONAL SYMPOSIUM ON SIGNAL PROCESSING AND ITS APPLICATIONS, VOL 1, PROCEEDINGS | 2003年
关键词：
D O I：
10.1109/ISSPA.2003.1224725
中图分类号：
TP18 [人工智能理论];
学科分类号：
081104 ; 0812 ; 0835 ; 1405 ;
摘要：
In Missing Feature Theory (MFT), noise robustness of speech recognizers is obtained by modifying the likelihood computed by the acoustic model to express that some features extracted from the signal are unreliable or missing. In one implementation of MFT, the acoustic model and bounds on the unreliable feature are used to infer an estimate of the missing data. This paper addresses an observed bias of the likelihood evaluated at the estimate. Theoretical and experimental evidence are provided that an upper bound on the accuracy is improved by applying a computationally simple correction for the number of free variables in the likelihood maximization.
引用
收藏
页码：401 / 404
页数：4
相关论文
共 50 条
  • [41] Robust Feature Combination for Speech Recognition Using Linear Microphone Array in a Car
    Obuchi, Yasunari
    Hataoka, Nobuo
    IN-VEHICLE CORPUS AND SIGNAL PROCESSING FOR DRIVER BEHAVIOR, 2009, : 187 - +
  • [42] Robust speech recognition with feature extraction using combined method of RSF and DRA
    Wada, N
    Hayasaka, N
    Yoshizawa, S
    Miyanaga, Y
    IEEE INTERNATIONAL SYMPOSIUM ON COMMUNICATIONS AND INFORMATION TECHNOLOGIES 2004 (ISCIT 2004), PROCEEDINGS, VOLS 1 AND 2: SMART INFO-MEDIA SYSTEMS, 2004, : 1001 - 1004
  • [43] 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):
  • [44] Missing data techniques using voicing probability dor robust automatic speech recognition
    Kim, LY
    Cho, HY
    Oh, YH
    ELECTRONICS LETTERS, 2001, 37 (11) : 723 - 724
  • [45] Robust speech recognition using cepstral domain Missing Data Techniques and noisy masks
    Van Hamme, H
    2004 IEEE INTERNATIONAL CONFERENCE ON ACOUSTICS, SPEECH, AND SIGNAL PROCESSING, VOL I, PROCEEDINGS: SPEECH PROCESSING, 2004, : 213 - 216
  • [46] Robust speech recognition using missing data techniques in the prospect domain and fuzzy masks
    Van Segbroeck, Maarten
    Van Hamme, Hugo
    2008 IEEE INTERNATIONAL CONFERENCE ON ACOUSTICS, SPEECH AND SIGNAL PROCESSING, VOLS 1-12, 2008, : 4393 - 4396
  • [47] Combining speech enhancement and auditory feature extraction for robust speech recognition
    Kleinschmidt, M
    Tchorz, J
    Kollmeier, B
    SPEECH COMMUNICATION, 2001, 34 (1-2) : 75 - 91
  • [48] MVDR based feature extraction for robust speech recognition
    Dharanipragada, S
    Rao, BD
    2001 IEEE INTERNATIONAL CONFERENCE ON ACOUSTICS, SPEECH, AND SIGNAL PROCESSING, VOLS I-VI, PROCEEDINGS: VOL I: SPEECH PROCESSING 1; VOL II: SPEECH PROCESSING 2 IND TECHNOL TRACK DESIGN & IMPLEMENTATION OF SIGNAL PROCESSING SYSTEMS NEURALNETWORKS FOR SIGNAL PROCESSING; VOL III: IMAGE & MULTIDIMENSIONAL SIGNAL PROCESSING MULTIMEDIA SIGNAL PROCESSING, 2001, : 309 - 312
  • [49] A Robust Feature Normalization Algorithm for Automatic Speech Recognition
    Lei, Jianjun
    Yang, Zhen
    Wang, Jian
    FIRST IITA INTERNATIONAL JOINT CONFERENCE ON ARTIFICIAL INTELLIGENCE, PROCEEDINGS, 2009, : 473 - +
  • [50] AN AUDITORY-BASED FEATURE FOR ROBUST SPEECH RECOGNITION
    Shao, Yang
    Jin, Zhaozhang
    Wang, DeLiang
    Srinivasan, Soundararajan
    2009 IEEE INTERNATIONAL CONFERENCE ON ACOUSTICS, SPEECH, AND SIGNAL PROCESSING, VOLS 1- 8, PROCEEDINGS, 2009, : 4625 - +
12345