Separation and deconvolution of speech using recurrent neural networks

被引：0
|
作者：
Li, Y [1 ]
Powers, D [1 ]
Wen, P [1 ]
机构：
[1] Flinders Univ S Australia, Sch Informat & Engn, Adelaide, SA 5001, Australia
来源：
IC-AI'2001: PROCEEDINGS OF THE INTERNATIONAL CONFERENCE ON ARTIFICIAL INTELLIGENCE, VOLS I-III | 2001年
关键词：
blind signal/source separation; speech recognition; recurrent neural networks; 2D system theory; output decorrelation;
D O I：
暂无
中图分类号：
TP18 [人工智能理论];
学科分类号：
081104 ; 0812 ; 0835 ; 1405 ;
摘要：
This paper focuses on improvements of the Speech Recognition or Speech Reading (SR), due to combining multiple auditory sources. We present results obtained in the traditional Blind Signal Separation & Deconvolution (BSS) paradigm using two speaker signals from the perspective of two sources, investigating artificial linear and convolutive mixes as well as real recordings. The adaptive algorithm is based on two-dimensional (2D) system theory using recurrent neural networks (RNNs). The characteristics of convolutively mixed signals (eg. audio signals) are matched by the structure of RNNs. The feedback paths in an RNN permit the possibility of a memory of the signals at relevant delays so that better separation can be achieved. The cross-correlations of the outputs of the RNN are used as separation criterion.
引用
收藏
页码：1303 / 1309
页数：7
相关论文
共 50 条
  • [1] Variational Recurrent Neural Networks for Speech Separation
    Chien, Jen-Tzung
    Kuo, Kuan-Ting
    [J]. 18TH ANNUAL CONFERENCE OF THE INTERNATIONAL SPEECH COMMUNICATION ASSOCIATION (INTERSPEECH 2017), VOLS 1-6: SITUATED INTERACTION, 2017, : 1193 - 1197
  • [2] Speech separation based on higher order statistics using Recurrent Neural Networks
    Li, Y
    Powers, DMW
    [J]. HYBRID INFORMATION SYSTEMS, 2002, : 45 - 55
  • [3] DISCRIMINATIVE DEEP RECURRENT NEURAL NETWORKS FOR MONAURAL SPEECH SEPARATION
    Wang, Guan-Xiang
    Hsu, Chung-Chien
    Chien, Jen-Tzung
    [J]. 2016 IEEE INTERNATIONAL CONFERENCE ON ACOUSTICS, SPEECH AND SIGNAL PROCESSING PROCEEDINGS, 2016, : 2544 - 2548
  • [4] RECURRENT NEURAL NETWORKS FOR COCHANNEL SPEECH SEPARATION IN REVERBERANT ENVIRONMENTS
    Delfarah, Masood
    Wang, DeLiang
    [J]. 2018 IEEE INTERNATIONAL CONFERENCE ON ACOUSTICS, SPEECH AND SIGNAL PROCESSING (ICASSP), 2018, : 5404 - 5408
  • [5] Speech prediction using recurrent neural networks
    Varoglu, E
    Hacioglu, K
    [J]. ELECTRONICS LETTERS, 1999, 35 (16) : 1353 - 1355
  • [6] Arabic speech recognition using recurrent neural networks
    El Choubassi, MM
    El Khoury, HE
    Alagha, CEJ
    Skaf, JA
    Al-Alaoui, MA
    [J]. PROCEEDINGS OF THE 3RD IEEE INTERNATIONAL SYMPOSIUM ON SIGNAL PROCESSING AND INFORMATION TECHNOLOGY, 2003, : 543 - 547
  • [7] PHASE-SENSITIVE AND RECOGNITION-BOOSTED SPEECH SEPARATION USING DEEP RECURRENT NEURAL NETWORKS
    Erdogan, Hakan
    Hershey, John R.
    Watanabe, Shinji
    Le Roux, Jonathan
    [J]. 2015 IEEE INTERNATIONAL CONFERENCE ON ACOUSTICS, SPEECH, AND SIGNAL PROCESSING (ICASSP), 2015, : 708 - 712
  • [8] Discriminatively Trained Recurrent Neural Networks for Single-Channel Speech Separation
    Weninger, Felix
    Hershey, John R.
    Le Roux, Jonathan
    Schuller, Bjoern
    [J]. 2014 IEEE GLOBAL CONFERENCE ON SIGNAL AND INFORMATION PROCESSING (GLOBALSIP), 2014, : 577 - 581
  • [9] Vietnamese Speech Command Recognition using Recurrent Neural Networks
    Phan Duy Hung
    Truong Minh Giang
    Le Hoang Nam
    Phan Minh Duong
    [J]. INTERNATIONAL JOURNAL OF ADVANCED COMPUTER SCIENCE AND APPLICATIONS, 2019, 10 (07) : 194 - 201
  • [10] Speech Emotion Recognition using Convolutional and Recurrent Neural Networks
    Lim, Wootaek
    Jang, Daeyoung
    Lee, Taejin
    [J]. 2016 ASIA-PACIFIC SIGNAL AND INFORMATION PROCESSING ASSOCIATION ANNUAL SUMMIT AND CONFERENCE (APSIPA), 2016,
12345