A KIND OF HYBRID LANGUAGE MODEL EMPLOYED IN POST PROCESSING OF CHINESE SPEECH RECOGNITION

被引：0
|
作者：
Mei Yong Xu Bingzheng Wang Qunsheng(Dept. of Radio Engineering
机构：
来源：
JournalofElectronics(China) | 1998年 / 03期
基金：
美国国家科学基金会;
关键词：
Trigram model; Tripos model; Combined model; Perplexity; Robustness;
D O I：
暂无
中图分类号：
TP391.9 [计算机仿真];
学科分类号：
080203 ;
摘要：
A kind of Chinese combined language model, that takes into account POS(part of speech) information in a trigram-based statistical language model, is presented in the paper. The theoretical analysis and experiments all show that the model not only is lower than trigram model in PP(perplexity), but also is superior to trigram model in dependence on testing text domain.
引用
收藏
页码：221 / 225
页数：5
相关论文
共 50 条
  • [41] A hybrid model for gesture recognition and speech synchronization
    Maniscalco, Umberto
    Messina, Antonio
    Storniolo, Pietro
    2023 IEEE INTERNATIONAL CONFERENCE ON ADVANCED ROBOTICS AND ITS SOCIAL IMPACTS, ARSO, 2023, : 27 - 32
  • [42] Combining speech enhancement with feature post-processing for robust speech recognition
    Lei, Jianjun
    Guo, Jun
    Liu, Gang
    Wang, Jian
    Nie, Xiangfei
    Yang, Zhen
    INTELLIGENT COMPUTING IN SIGNAL PROCESSING AND PATTERN RECOGNITION, 2006, 345 : 773 - 778
  • [43] Manchu handwritten character recognition post-processing based on the combination language model
    Zhang, GY
    Li, JJ
    Wang, AX
    PROCEEDINGS OF THE 2005 IEEE INTERNATIONAL CONFERENCE ON NATURAL LANGUAGE PROCESSING AND KNOWLEDGE ENGINEERING (IEEE NLP-KE'05), 2005, : 42 - 46
  • [44] Recognition of Chinese speech using hybrid HMM/HNN models
    Jia, Ying
    Du, Limin
    Hou, Ziqiang
    International Conference on Signal Processing Proceedings, ICSP, 1998, 1 : 726 - 729
  • [45] Recognition of Chinese speech using hybrid HMM HNN models
    Jia, Y
    Du, LM
    Hou, ZQ
    ICSP '98: 1998 FOURTH INTERNATIONAL CONFERENCE ON SIGNAL PROCESSING, PROCEEDINGS, VOLS I AND II, 1998, : 726 - 729
  • [46] Chinese Speech Recognition Based on a Hybrid SVM and HMM Architecture
    Luo, Xingxian
    ADVANCES IN NEURAL NETWORKS - ISNN 2011, PT III, 2011, 6677 : 629 - 635
  • [47] Speech recognition based on unified model of acoustic and language aspects of speech
    1600, Nippon Telegraph and Telephone Corp. (11):
  • [48] Influence of language models and candidate set size on contextual post-processing for Chinese script recognition
    Li, YX
    Tan, CL
    PROCEEDINGS OF THE 17TH INTERNATIONAL CONFERENCE ON PATTERN RECOGNITION, VOL 2, 2004, : 537 - 540
  • [49] Towards improving speech recognition model with post-processing spell correction using BERT
    Priya, M. C. Shunmuga
    Renuka, D. Karthika
    Kumar, L. Ashok
    JOURNAL OF INTELLIGENT & FUZZY SYSTEMS, 2022, 43 (04) : 4873 - 4882
  • [50] MULTI-CLASS AUTOMATED SPEECH LANGUAGE RECOGNITION USING NATURAL LANGUAGE PROCESSING WITH OPTIMAL DEEP LEARNING MODEL
    Al-anazi, Reema g.
    Alqahtani, Hamed
    Alzaidi, Muhammad swaileh a.
    Alanazi, Meshari h.
    AL Sultan, Hanan
    Alrowaily, Amal f.
    Aljabri, Jawhara
    Alqudah, Assal
    FRACTALS-COMPLEX GEOMETRY PATTERNS AND SCALING IN NATURE AND SOCIETY, 2025,
12345