The XMUSPEECH System for Accented English Automatic Speech Recognition

被引：2
|
作者：
Tong, Fuchuan [1 ]
Li, Tao [2 ]
Liao, Dexin [2 ]
Xia, Shipeng [2 ]
Li, Song [1 ]
Hong, Qingyang [2 ]
Li, Lin [1 ]
机构：
[1] Xiamen Univ, Sch Elect Sci & Engn, Xiamen 361005, Peoples R China
[2] Xiamen Univ, Sch Informat, Xiamen 361005, Peoples R China
来源：
APPLIED SCIENCES-BASEL | 2022年 / 12卷 / 03期
基金：
中国国家自然科学基金;
关键词：
AESRC2020; i-vector; x-vector; multistream CNN;
D O I：
10.3390/app12031478
中图分类号：
O6 [化学];
学科分类号：
0703 ;
摘要：
In this paper, we present the XMUSPEECH systems for Track 2 of the Interspeech 2020 Accented English Speech Recognition Challenge (AESRC2020). Track 2 is an Automatic Speech Recognition (ASR) task where the non-native English speakers have various accents, which reduces the accuracy of the ASR system. To solve this problem, we experimented with acoustic models and input features. Furthermore, we trained a TDNN-LSTM language model for lattice rescoring to obtain better results. Compared with our baseline system, we achieved relative word error rate (WER) improvements of 40.7% and 35.7% on the development set and evaluation set, respectively.
引用
收藏
页数：8
相关论文
共 50 条
  • [1] Automatic Speech Recognition of Multiple Accented English Data
    Vergyri, Dimitra
    Lamel, Lori
    Gauvain, Jean-Luc
    11TH ANNUAL CONFERENCE OF THE INTERNATIONAL SPEECH COMMUNICATION ASSOCIATION 2010 (INTERSPEECH 2010), VOLS 3 AND 4, 2010, : 1652 - +
  • [2] Inequity in Popular Speech Recognition Systems for Accented English Speech
    Ike, Chinaemere
    Polsley, Seth
    Hammond, Tracy
    COMPANION PROCEEDINGS OF THE 27TH INTERNATIONAL CONFERENCE ON INTELLIGENT USER INTERFACES, IUI 2022 COMPANION, 2022, : 66 - 68
  • [3] AISPEECH-SJTU ASR SYSTEM FOR THE ACCENTED ENGLISH SPEECH RECOGNITION CHALLENGE
    Tan, Tian
    Lu, Yizhou
    Ma, Rao
    Zhu, Sen
    Guo, Jiaqi
    Qian, Yanmin
    2021 IEEE INTERNATIONAL CONFERENCE ON ACOUSTICS, SPEECH AND SIGNAL PROCESSING (ICASSP 2021), 2021, : 6413 - 6417
  • [4] Automatic accent classification of foreign accented Australian English speech
    Kumpf, K
    King, RW
    ICSLP 96 - FOURTH INTERNATIONAL CONFERENCE ON SPOKEN LANGUAGE PROCESSING, PROCEEDINGS, VOLS 1-4, 1996, : 1740 - 1743
  • [5] AISPEECH-SJTU ACCENT IDENTIFICATION SYSTEM FOR THE ACCENTED ENGLISH SPEECH RECOGNITION CHALLENGE
    Huang, Houjun
    Xiang, Xu
    Yang, Yexin
    Ma, Rao
    Qian, Yanmin
    2021 IEEE INTERNATIONAL CONFERENCE ON ACOUSTICS, SPEECH AND SIGNAL PROCESSING (ICASSP 2021), 2021, : 6254 - 6258
  • [6] Robust Automatic Speech Recognition for Accented Mandarin in Car Environments
    Pei Ding
    Lei He
    Xiang Yan
    Jie Hao
    INTERSPEECH 2006 AND 9TH INTERNATIONAL CONFERENCE ON SPOKEN LANGUAGE PROCESSING, VOLS 1-5, 2006, : 2542 - 2545
  • [7] Recognition of foreign-accented vocoded speech by native English listeners
    Yang, Jing
    Barrett, Jenna
    Yin, Zhigang
    Xu, Li
    ACTA ACUSTICA, 2023, 7
  • [8] EVALUATIONAL REACTIONS TO ACCENTED ENGLISH SPEECH
    ANISFELD, M
    LAMBERT, WE
    BOGO, N
    JOURNAL OF ABNORMAL PSYCHOLOGY, 1962, 65 (04) : 223 - &
  • [9] SPEECH RECOGNITION OF MULTIPLE ACCENTED ENGLISH DATA USING ACOUSTIC MODEL INTERPOLATION
    Fraga-Silva, Thiago
    Gauvain, Jean-Luc
    Lamel, Lori
    2014 PROCEEDINGS OF THE 22ND EUROPEAN SIGNAL PROCESSING CONFERENCE (EUSIPCO), 2014, : 1781 - 1785
  • [10] Development of HMM Based Automatic Speech Recognition System For Indian English
    Garud, Anushri
    Bang, Arti
    Joshi, Shrikant
    2018 FOURTH INTERNATIONAL CONFERENCE ON COMPUTING COMMUNICATION CONTROL AND AUTOMATION (ICCUBEA), 2018,
12345