Cross-Corpora Language Recognition: A Preliminary Investigation with Indian Languages

被引:7
|
作者
Dey, Spandan [1 ]
Saha, Goutam [1 ]
Sahidullah, Md [2 ]
机构
[1] Indian Inst Technol, Dept E&ECE, Kharagpur, W Bengal, India
[2] Univ Lorraine, CNRS, INRIA, LORIA, F-54000 Nancy, France
关键词
Cross-corpora; language recognition; channel compensation; long-term average spectrum; TDNN; IDENTIFICATION;
D O I
10.23919/EUSIPCO54536.2021.9616273
中图分类号
O42 [声学];
学科分类号
070206 ; 082403 ;
摘要
In this paper, we conduct one of the very first studies for cross-corpora performance evaluation in the spoken language identification (LID) problem. Cross-corpora evaluation was not explored much in LID research, especially for the Indian languages. We have selected three Indian spoken language corpora: IIITH-ILSC, LDC South Asian, and IITKGP-MLILSC. For each of the corpus, LID systems are trained on the state-of-the-art time-delay neural network (TDNN) based architecture with MFCC features. We observe that the LID performance degrades drastically for cross-corpora evaluation. For example, the system trained on the IIITH-ILSC corpus shows an average EER of 11.80 % and 43.34 % when evaluated with the same corpora and LDC South Asian corpora, respectively. Our preliminary analysis shows the significant differences among these corpora in terms of mismatch in the long-term average spectrum (LTAS) and signal-to-noise ratio (SNR). Subsequently, we apply different feature level compensation methods to reduce the cross-corpora acoustic mismatch. Our results indicate that these feature normalization schemes can help to achieve promising LID performance on cross-corpora experiments.
引用
收藏
页码:546 / 550
页数:5
相关论文
共 50 条
  • [41] A hierarchical language identification system for Indian languages
    Jothilakshmi, S.
    Ramalingam, V.
    Palanivel, S.
    DIGITAL SIGNAL PROCESSING, 2012, 22 (03) : 544 - 553
  • [42] INDIAN SIGN LANGUAGE RECOGNITION
    Deora, Divya
    Bajaj, Nikesh
    2012 1ST INTERNATIONAL CONFERENCE ON EMERGING TECHNOLOGY TRENDS IN ELECTRONICS, COMMUNICATION AND NETWORKING (ET2ECN), 2012,
  • [43] Speech Emotion Recognition Cross Language Families: Mandarin vs. Western Languages
    Xiao, Zhongzhe
    Wu, Di
    Zhang, Xiaojun
    Tao, Zhi
    PROCEEDINGS OF THE 2016 INTERNATIONAL CONFERENCE ON PROGRESS IN INFORMATICS AND COMPUTING (PIC), VOL 1, 2016, : 253 - 257
  • [44] CROSS-LINGUAL SPEECH RECOGNITION BETWEEN LANGUAGES FROM THE SAME LANGUAGE FAMILY
    Zgank, Andrej
    PROCEEDINGS OF THE ROMANIAN ACADEMY SERIES A-MATHEMATICS PHYSICS TECHNICAL SCIENCES INFORMATION SCIENCE, 2019, 20 (02): : 184 - 191
  • [45] Collecting and evaluating speech recognition corpora for 11 South African languages
    Badenhorst, Jaco
    van Heerden, Charl
    Davel, Marelie
    Barnard, Etienne
    LANGUAGE RESOURCES AND EVALUATION, 2011, 45 (03) : 289 - 309
  • [46] Collecting and evaluating speech recognition corpora for 11 South African languages
    Jaco Badenhorst
    Charl van Heerden
    Marelie Davel
    Etienne Barnard
    Language Resources and Evaluation, 2011, 45 : 289 - 309
  • [47] Investigation of automatic mixed-lingual affective state recognition system for diverse Indian languages
    Lalitha, S.
    Gupta, Deepa
    JOURNAL OF INTELLIGENT & FUZZY SYSTEMS, 2021, 41 (05) : 5467 - 5476
  • [48] Investigation of automatic mixed-lingual affective state recognition system for diverse Indian languages
    Lalitha, S.
    Gupta, Deepa
    Journal of Intelligent and Fuzzy Systems, 2021, 41 (05): : 5467 - 5476
  • [49] A Preliminary Cross-Language Investigation of Empathy in # Disability Discourse on Twitter
    AlMeraj, Zainab
    Husain, Fatemah
    PROCEEDINGS OF THE 16TH ACM INTERNATIONAL CONFERENCE ON PERVASIVE TECHNOLOGIES RELATED TO ASSISTIVE ENVIRONMENTS, PETRA 2023, 2023, : 675 - 679
  • [50] Development of Multilingual Phone Recognition System for Indian Languages
    Manjunath, K. E.
    Rao, K. Sreenivasa
    Jayagopi, Dinesh Babu
    2017 IEEE INTERNATIONAL CONFERENCE ON SIGNAL PROCESSING, INFORMATICS, COMMUNICATION AND ENERGY SYSTEMS (SPICES), 2017,