LEARNABLE NONLINEAR COMPRESSION FOR ROBUST SPEAKER VERIFICATION

被引：2
|
作者：
Liu, Xuechen [1 ,2 ]
Sahidullah, Md [2 ]
Kinnunen, Tomi [1 ]
机构：
[1] Univ Eastern Finland, Sch Comp, Joensuu, Finland
[2] Univ Lorraine, INRIA, CNRS, LORIA, F-54000 Nancy, France
来源：
2022 IEEE INTERNATIONAL CONFERENCE ON ACOUSTICS, SPEECH AND SIGNAL PROCESSING (ICASSP) | 2022年
关键词：
Speaker Verification; Nonlinear Compression; Multi-Regime Compression; RECOGNITION;
D O I：
10.1109/ICASSP43922.2022.9747185
中图分类号：
O42 [声学];
学科分类号：
070206 ; 082403 ;
摘要：
In this study, we focus on nonlinear compression methods in spectral features for speaker verification based on deep neural network. We consider different kinds of channel-dependent (CD) nonlinear compression methods optimized in a data-driven manner. Our methods are based on power nonlinearities and dynamic range compression (DRC). We also propose multi-regime (MR) design on the nonlinearities, at improving robustness. Results on VoxCeleb1 and VoxMovies data demonstrate improvements brought by proposed compression methods over both the commonly-used logarithm and their static counterparts, especially for ones based on power function. While CD generalization improves performance on VoxCeleb1, MR provides more robustness on VoxMovies, with a maximum relative equal error rate reduction of 21.6%.
引用
收藏
页码：7962 / 7966
页数：5
相关论文
共 50 条
  • [41] Channel robust speaker verification via feature mapping
    Reynolds, DA
    2003 IEEE INTERNATIONAL CONFERENCE ON ACOUSTICS, SPEECH, AND SIGNAL PROCESSING, VOL II, PROCEEDINGS: SPEECH II; INDUSTRY TECHNOLOGY TRACKS; DESIGN & IMPLEMENTATION OF SIGNAL PROCESSING SYSTEMS; NEURAL NETWORKS FOR SIGNAL PROCESSING, 2003, : 53 - 56
  • [42] Senone I-Vectors for Robust Speaker Verification
    Tan, Zhili
    Zhu, Yingke
    Mak, Man-Wai
    Mak, Brian Kan-Wing
    2016 10TH INTERNATIONAL SYMPOSIUM ON CHINESE SPOKEN LANGUAGE PROCESSING (ISCSLP), 2016,
  • [43] Robust speaker verification over the telephone by feature recuperation
    Li, X
    Mak, MW
    Kung, SY
    PROCEEDINGS OF 2001 INTERNATIONAL SYMPOSIUM ON INTELLIGENT MULTIMEDIA, VIDEO AND SPEECH PROCESSING, 2001, : 433 - 436
  • [44] Robust Training for Speaker Verification against Noisy Labels
    Fang, Zhihua
    He, Liang
    Ma, Hanhan
    Guo, Xiaochen
    Li, Lin
    INTERSPEECH 2023, 2023, : 3192 - 3196
  • [45] Kurtosis normalization in feature space for robust speaker verification
    Xie, Yanlu
    Dai, Beiqian
    Yao, Zhiqiang
    Liu, Minghui
    2006 IEEE International Conference on Acoustics, Speech and Signal Processing, Vols 1-13, 2006, : 117 - 120
  • [46] A speaker verification backend with robust performance across conditions
    Ferrer, Luciana
    McLaren, Mitchell
    Brummer, Niko
    COMPUTER SPEECH AND LANGUAGE, 2022, 71
  • [47] Noise Robust Speaker Verification with Delta Cepstrum Normalization
    Kanda, Naoyuki
    Takeda, Ryu
    Obuchi, Yasunari
    14TH ANNUAL CONFERENCE OF THE INTERNATIONAL SPEECH COMMUNICATION ASSOCIATION (INTERSPEECH 2013), VOLS 1-5, 2013, : 3111 - 3115
  • [48] Robust speaker verification in low bit rate channels
    Altinçay, H
    Ergün, C
    Ahmad, W
    ELECTRONICS LETTERS, 2003, 39 (06) : 576 - 578
  • [49] Probabilistic latent prosody analysis for robust speaker verification
    Chen, Zi-He
    Zeng, Zhi-Ren
    Liao, Yuan-Fu
    Juang, Yau-Tarng
    2006 IEEE INTERNATIONAL CONFERENCE ON ACOUSTICS, SPEECH AND SIGNAL PROCESSING, VOLS 1-13, 2006, : 105 - 108
  • [50] MODELLING SPEAKER AND CHANNEL VARIABILITY USING DEEP NEURAL NETWORKS FOR ROBUST SPEAKER VERIFICATION
    Bhattacharya, Gautam
    Alam, Jahangir
    Kenny, Patrick
    Gupta, Vishwa
    2016 IEEE WORKSHOP ON SPOKEN LANGUAGE TECHNOLOGY (SLT 2016), 2016, : 192 - 198
12345