ROBUST SPEECH RECOGNITION USING MULTIVARIATE COPULA MODELS

被引：0
|
作者：
Bayestehtashk, Alireza [1 ]
Shafran, Izhak [2 ]
Babaeian, Amir [3 ]
机构：
[1] Oregon Hlth & Sci Univ, Portland, OR 97201 USA
[2] Google Inc, Mountain View, CA USA
[3] Univ Calif San Diego, La Jolla, CA 92093 USA
来源：
2016 IEEE INTERNATIONAL CONFERENCE ON ACOUSTICS, SPEECH AND SIGNAL PROCESSING PROCEEDINGS | 2016年
关键词：
Copula model; Robust speech recognition; Deep neural network; Aurora; 4;
D O I：
暂无
中图分类号：
O42 [声学];
学科分类号：
070206 ; 082403 ;
摘要：
In this paper, we continue our investigation into copula models for real-valued multivariate features with the goal of compensating for the mismatch in the training and the testing conditions. Previously, we reported results on UCI classification tasks where our method consistently outperformed other competing classifiers [1]. Here, we extend this work from classification to recognition and elaborate further on the mathematical properties of our models in the form of lemmas. We report results on the Aurora 4 automatic speech recognition (ASR) task which contains utterances with wide range of background noise that are not well represented in the training data. Our results show that the proposed copula-based models improve the accuracy by about 7% (11.6 vs 12.4) over a comparable baseline.
引用
收藏
页码：5890 / 5894
页数：5
相关论文
共 50 条
  • [31] Limited training data robust speech recognition using kernel-based acoustic models
    Schaffoener, Martin
    Krueger, Sven E.
    Andelic, Edin
    Katz, Marcel
    Wendemuth, Andreas
    2006 IEEE International Conference on Acoustics, Speech and Signal Processing, Vols 1-13, 2006, : 1137 - 1140
  • [32] ALGONQUIN - Learning dynamic noise models from noisy speech for robust speech recognition
    Frey, BJ
    Kristjansson, TT
    Deng, L
    Acero, A
    ADVANCES IN NEURAL INFORMATION PROCESSING SYSTEMS 14, VOLS 1 AND 2, 2002, 14 : 1165 - 1171
  • [33] Towards Robust Indonesian Speech Recognition with Spontaneous-Speech Adapted Acoustic Models
    Hoesen, Devin
    Satriawan, Cil Hardianto
    Lestari, Dessi Puji
    Khodra, Masayu Leylia
    SLTU-2016 5TH WORKSHOP ON SPOKEN LANGUAGE TECHNOLOGIES FOR UNDER-RESOURCED LANGUAGES, 2016, 81 : 167 - 173
  • [34] A study of robust speech recognition using FRM filter
    Hayasaka, N
    Miyanaga, Y
    TENCON 2004 - 2004 IEEE REGION 10 CONFERENCE, VOLS A-D, PROCEEDINGS: ANALOG AND DIGITAL TECHNIQUES IN ELECTRICAL ENGINEERING, 2004, : A80 - A83
  • [35] ROBUST SPEECH RECOGNITION USING GENERATIVE ADVERSARIAL NETWORKS
    Sriram, Anuroop
    Jun, Heewoo
    Gaur, Yashesh
    Satheesh, Sanjeev
    2018 IEEE INTERNATIONAL CONFERENCE ON ACOUSTICS, SPEECH AND SIGNAL PROCESSING (ICASSP), 2018, : 5639 - 5643
  • [36] Robust speech recognition using a noise rejection approach
    Khan, E
    Levinson, R
    IEEE INTERNATIONAL JOINT SYMPOSIA ON INTELLIGENCE AND SYSTEMS - PROCEEDINGS, 1998, : 326 - 335
  • [37] Robust Speech Recognition using Generalized Distillation Framework
    Markov, Konstantin
    Matsui, Tomoko
    17TH ANNUAL CONFERENCE OF THE INTERNATIONAL SPEECH COMMUNICATION ASSOCIATION (INTERSPEECH 2016), VOLS 1-5: UNDERSTANDING SPEECH PROCESSING IN HUMANS AND MACHINES, 2016, : 2364 - 2368
  • [38] Robust speech recognition by using compensated acoustic scores
    Sato, S
    Onoe, K
    Kobayashi, A
    Imai, T
    IEICE TRANSACTIONS ON INFORMATION AND SYSTEMS, 2006, E89D (03): : 915 - 921
  • [39] Robust speech recognition using wavelet coefficient features
    Gupta, M
    Gilbert, A
    ASRU 2001: IEEE WORKSHOP ON AUTOMATIC SPEECH RECOGNITION AND UNDERSTANDING, CONFERENCE PROCEEDINGS, 2001, : 445 - 448
  • [40] ROBUST ISOLATED SPEECH RECOGNITION USING BINARY MASKS
    Karadogan, Seliz Gulsen
    Larsen, Jan
    Pedersen, Michael Syskind
    Boldt, Jesper Bunsow
    18TH EUROPEAN SIGNAL PROCESSING CONFERENCE (EUSIPCO-2010), 2010, : 1988 - 1992
12345