Uyghur continuous speech recognition based on HTK

被引：0
|
作者：
Tursun, Nasirjan [1 ]
Silamu, Wushour [1 ]
Tao, Mei [1 ]
机构：
[1] Xinjiang Univ, Sch Math & Sys Sci, Urumqi 830046, Peoples R China
来源：
RECENT ADVANCE OF CHINESE COMPUTING TECHNOLOGIES | 2007年
关键词：
Uyghur language; continuous speech recognition; HTK;
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：
We present the process of Uyghur continuous speech recognition and an experiment on Uyghur continuous speech recognition using HTK. In this paper, we choose the monophone and the triphone as recognition unit for acoustic model training, and the statistical 2-gram model for the language model. To increase recognition accuracy, we also optimize the acoustic model by increasing the Gaussian mixture components and bundling the model state. This experiment is based on the SRSD (Speech Recognition Speech Database for Uyghur), which includes 64 speakers.
引用
收藏
页码：377 / 380
页数：4
相关论文
共 50 条
  • [1] Syllable based language model for large vocabulary continuous speech recognition of Uyghur
    Silamu, W. (wushour@xju.edu.cn), 1600, Tsinghua University (53):
  • [2] HTK-Based Recognition of Whispered Speech
    Galic, Jovan
    Jovicic, Slobodan T.
    Grozdic, Dorde
    Markovic, Branko
    SPEECH AND COMPUTER, 2014, 8773 : 251 - 258
  • [3] Research on speech recognition algorithm based on HTK toolbox
    Wang, Lei
    PROCEEDINGS OF THE 2016 3RD INTERNATIONAL CONFERENCE ON MATERIALS ENGINEERING, MANUFACTURING TECHNOLOGY AND CONTROL, 2016, 67 : 184 - 186
  • [4] Large Vocabulary Continuous Speech Recognition of Uyghur: Basic Research of Decoder
    Shadike, Muhetaer
    Li, Xiao
    Wasili, Buheliqiguli
    ADVANCES IN NEURAL NETWORKS - ISNN 2011, PT III, 2011, 6677 : 594 - +
  • [5] Large Vocabulary Continuous Speech Recognition in Uyghur: Data Preparation and Experimental Results
    Tursun, Nasirjan
    Silamu, Wushour
    2008 6TH INTERNATIONAL SYMPOSIUM ON CHINESE SPOKEN LANGUAGE PROCESSING, PROCEEDINGS, 2008, : 197 - 200
  • [6] Using HTK, ISIP and Julius decoders in Slovenian large vocabulary continuous speech recognition
    Rotovnik, Tomaž
    Horvat, Bogomir
    Kačič, Zdravko
    Elektrotehniski Vestnik/Electrotechnical Review, 2002, 69 (05): : 253 - 258
  • [7] A Window-based Assistant Tool for HTK Speech Recognition Research
    Su, Jie
    Guo, Lin
    Li, Yu-fang
    Liu, Jing-feng
    Pan, Wen-lin
    2015 INTERNATIONAL CONFERENCE ON APPLIED MECHANICS AND MECHATRONICS ENGINEERING (AMME 2015), 2015, : 579 - 584
  • [8] A MFCC based Hindi Speech Recognition Technique using HTK Toolkit
    Tripathy, Shweta
    Baranwal, Neha
    Nandi, G. C.
    2013 IEEE SECOND INTERNATIONAL CONFERENCE ON IMAGE INFORMATION PROCESSING (ICIIP), 2013, : 539 - 544
  • [9] Punjabi automatic speech recognition using HTK
    Dua, Mohit
    Aggarwal, R.K.
    Kadyan, Virender
    Dua, Shelza
    International Journal of Computer Science Issues, 2012, 9 (4 4-1): : 359 - 364
  • [10] Morpheme-based Uyghur speech recognition considering vowel weakening
    Xue, Huajian
    Zhang, Ronghui
    Osman, Turghun
    Li, Xiao
    International Journal of Advancements in Computing Technology, 2012, 4 (01) : 289 - 296
12345