Online directional speech enhancement using geometrically constrained independent vector analysis

被引:11
|
作者
Li, Li [1 ]
Koishida, Kazuhito [2 ]
Makino, Shoji [1 ]
机构
[1] Univ Tsukuba, Tsukuba, Ibaraki, Japan
[2] Microsoft Corp, Redmond, WA 98052 USA
来源
INTERSPEECH 2020 | 2020年
关键词
multichannel speech enhancement; geometric constraint; independent vector analysis (IVA); online; real-time; BLIND SOURCE SEPARATION; PERMUTATION PROBLEM; ICA;
D O I
10.21437/Interspeech.2020-1484
中图分类号
R36 [病理学]; R76 [耳鼻咽喉科学];
学科分类号
100104 ; 100213 ;
摘要
This paper proposes an online dual-microphone system for directional speech enhancement, which employs geometrically constrained independent vector analysis (IVA) based on the auxiliary function approach and vectorwise coordinate descent. Its offline version has recently been proposed and shown to outperform the conventional auxiliary function approach-based IVA (AuxIVA) thanks to the properly designed spatial constraints. We extend the offline algorithm to online by incorporating the autoregressive approximation of an auxiliary variable. Experimental evaluations revealed that the proposed online algorithm could work in real-time and achieved superior speech enhancement performance to online AuxIVA in both situations where a fixed target was interfered by a spatially stationary or dynamic interference.
引用
收藏
页码:61 / 65
页数:5
相关论文
共 50 条
  • [1] GEOMETRICALLY CONSTRAINED INDEPENDENT VECTOR ANALYSIS FOR DIRECTIONAL SPEECH ENHANCEMENT
    Li, Li
    Koishida, Kazuhito
    2020 IEEE INTERNATIONAL CONFERENCE ON ACOUSTICS, SPEECH, AND SIGNAL PROCESSING, 2020, : 846 - 850
  • [2] Accelerating online algorithm using geometrically constrained independent vector analysis with iterative source steering
    Goto, Kana
    Ueda, Tetsuya
    Li, Li
    Yamada, Takeshi
    Makino, Shoji
    PROCEEDINGS OF 2022 ASIA-PACIFIC SIGNAL AND INFORMATION PROCESSING ASSOCIATION ANNUAL SUMMIT AND CONFERENCE (APSIPA ASC), 2022, : 754 - 759
  • [3] Geometrically Constrained Independent Vector Analysis with Auxiliary Function Approach and Iterative Source Steering
    Goto, Kana
    Ueda, Tetsuya
    Li, Li
    Yamada, Takeshi
    Makino, Shoji
    2022 30TH EUROPEAN SIGNAL PROCESSING CONFERENCE (EUSIPCO 2022), 2022, : 757 - 761
  • [4] Scale-certainty geometrically constrained independent vector analysis for determined blind source separation
    Gu, Jianjun
    Yao, Dingding
    Li, Junfeng
    Yan, Yonghong
    APPLIED ACOUSTICS, 2023, 206
  • [5] Efficient online target speech extraction using DOA-constrained independent component analysis of stereo data for robust speech recognition
    Kim, Minook
    Park, Hyung-Min
    SIGNAL PROCESSING, 2015, 117 : 126 - 137
  • [6] Audiovisual Speech Separation Based on Independent Vector Analysis Using a Visual Voice Activity Detector
    Narvor, Pierre
    Rivet, Bertrand
    Jutten, Christian
    LATENT VARIABLE ANALYSIS AND SIGNAL SEPARATION (LVA/ICA 2017), 2017, 10169 : 247 - 257
  • [7] NUMERICAL INVESTIGATION OF WEIGHT PARAMETERS FOR GEOMETRICALLY CONSTRAINED INDEPENDENT VECTOR ANALYSIS USING VECTORWISE COORDINATE DESCENT OR ITERATIVE SOURCE STEERING
    Furunaga, Shinya
    Goto, Kana
    Ueda, Tetsuya
    Li, Li
    Yamada, Takeshi
    Makino, Shoji
    2022 INTERNATIONAL WORKSHOP ON ACOUSTIC SIGNAL ENHANCEMENT (IWAENC 2022), 2022,
  • [8] Independent vector analysis for real world speech processing
    Lee, Intae
    Lee, Te-Won
    INDEPENDENT COMPONENT ANALYSES, WAVELETS, UNSUPERVISED NANO-BIOMIMETIC SENSORS, AND NEURAL NETWORKS V, 2007, 6576
  • [9] Independent vector analysis followed by HMM-based feature enhancement for robust speech recognition
    Cho, Ji-Won
    Park, Hyung-Min
    SIGNAL PROCESSING, 2016, 120 : 200 - 208
  • [10] Bayesian feature enhancement using independent vector analysis and reverberation parameter re-estimation for noisy reverberant speech recognition
    Cho, Ji-Won
    Park, Jong-Hyeon
    Chang, Joon-Hyuk
    Park, Hyung-Min
    COMPUTER SPEECH AND LANGUAGE, 2017, 46 : 496 - 516