Prosodic Features for a Maximum Entropy Language Model

被引：0
|
作者：
Chan, Oscar [1 ]
Togneri, Roberto [1 ]
机构：
[1] Univ Western Australia, Sch Elect Elect & Comp Engn, Nedlands, WA 6009, Australia
来源：
INTERSPEECH 2006 AND 9TH INTERNATIONAL CONFERENCE ON SPOKEN LANGUAGE PROCESSING, VOLS 1-5 | 2006年
关键词：
language modelling; prosody; speech recognition;
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：
This paper presents an approach for incorporating prosodic knowledge into the language modelling component of a speech recogniser. We formulate features for a maximum entropy language model which capture various aspects of the relationships between prosody, syntax and the spoken word sequence. Maximum entropy is a powerful modelling technique, and well suited to modelling prosodic information. Tests conducted on the Boston University Radio Speech Corpus using this model showed improvements in perplexity, and n-best rescoring results also demonstrated small but statistically significant gains.
引用
收藏
页码：1858 / 1861
页数：4
相关论文
共 50 条
  • [1] Improvement of a whole sentence maximum entropy language model using grammatical features
    Amaya, F
    Benedi, JM
    39TH ANNUAL MEETING OF THE ASSOCIATION FOR COMPUTATIONAL LINGUISTICS, PROCEEDINGS OF THE CONFERENCE, 2001, : 10 - 17
  • [2] An improved maximum entropy language model
    Fang, GL
    Wen, G
    2002 6TH INTERNATIONAL CONFERENCE ON SIGNAL PROCESSING PROCEEDINGS, VOLS I AND II, 2002, : 1083 - 1086
  • [3] Prosodic features for language identification
    Mary, Leena
    Yegnanarayana, B.
    ICSCN 2008: PROCEEDINGS OF THE INTERNATIONAL CONFERENCE ON SIGNAL PROCESSING COMMUNICATIONS AND NETWORKING, 2008, : 57 - +
  • [4] A whole sentence maximum entropy language model
    Rosenfeld, R
    1997 IEEE WORKSHOP ON AUTOMATIC SPEECH RECOGNITION AND UNDERSTANDING, PROCEEDINGS, 1997, : 230 - 237
  • [5] Using continuous features in the maximum entropy model
    Yu, Dong
    Deng, Li
    Acero, Alex
    PATTERN RECOGNITION LETTERS, 2009, 30 (14) : 1295 - 1300
  • [6] A maximum entropy Markov model for prediction of prosodic phrase boundaries in Chinese TTS
    Zhao, Ziping
    Zhao, Tingjian
    Zhu, Yaoting
    GRC: 2007 IEEE INTERNATIONAL CONFERENCE ON GRANULAR COMPUTING, PROCEEDINGS, 2007, : 498 - 501
  • [7] A MAXIMUM ENTROPY BASED HIERARCHICAL MODEL FOR AUTOMATIC PROSODIC BOUNDARY LABELING IN MANDARIN
    Liu, Fangzhou
    Jia, Huibin
    Tao, Jianhua
    2008 6TH INTERNATIONAL SYMPOSIUM ON CHINESE SPOKEN LANGUAGE PROCESSING, PROCEEDINGS, 2008, : 257 - 260
  • [8] Using Dependency Grammar Features in Whole Sentence Maximum Entropy Language Model for Speech Recognition
    Ruokolainen, Teemu
    Alumaee, Tanel
    Dobrinkat, Marcus
    HUMAN LANGUAGE TECHNOLOGIES - THE BALTIC PERSPECTIVE, 2010, 219 : 73 - 79
  • [9] Spoken Language Recognition With Prosodic Features
    Ng, Raymond W. M.
    Lee, Tan
    Leung, Cheung-Chi
    Ma, Bin
    Li, Haizhou
    IEEE TRANSACTIONS ON AUDIO SPEECH AND LANGUAGE PROCESSING, 2013, 21 (09): : 1841 - 1853
  • [10] Improved maximum entropy language model and its application
    Li, Juanzi
    Huang, Changning
    Ruan Jian Xue Bao/Journal of Software, 1999, 10 (03): : 257 - 263
12345