Heuristic Improvements of the HMM Use in Isolated Word Speech Recognition

被引：0
作者：
Dimov, Dimo [1 ]
Azmanov, Ivan [1 ]
机构：
[1] Inst Informat Technol, Sofia 1113, Bulgaria
来源：
CYBERNETICS AND INFORMATION TECHNOLOGIES | 2007年 / 7卷 / 03期
关键词：
Speech recognition; HMM; Gaussian pdf; speech cepstrum;
D O I：
暂无
中图分类号：
TP [自动化技术、计算机技术];
学科分类号：
0812 ;
摘要：
The paper presents the authors' experience with HMMs (Hidden Markov Models) used for isolated word speech recognition in Bulgarian. Two methods provoked by experiments are discussed, namely: (i) the precise quantization of Gaussian probability density function (G-pdf) modeling the HMM states' output, and (ii) a method for averaging a set of HMMs trained for different versions of a given word. A universal threshold is evaluated for switching between the "large" and "slim" models of G-pdf defined here. Experimental results of the threshold usage for averaging of HMMs are also reported.
引用
收藏
页码：73 / 88
页数：16
相关论文
共 50 条
  • [31] The application of Speech Recognition Technology based on HMM
    Yan, Guilin
    PROCEEDINGS OF THE FIRST INTERNATIONAL CONFERENCE ON INFORMATION SCIENCES, MACHINERY, MATERIALS AND ENERGY (ICISMME 2015), 2015, 126 : 676 - 679
  • [32] Speech recognition with HMM models for cochlear prostheses
    Sakka, Z
    Kachouri, A
    Samet, M
    2004 IEEE International Conference on Industrial Technology (ICIT), Vols. 1- 3, 2004, : 1478 - 1481
  • [33] Factor Analyzed HMM Topology for Speech Recognition
    Ting, Chuan-Wei
    Chien, Jen-Tzung
    INTERSPEECH 2009: 10TH ANNUAL CONFERENCE OF THE INTERNATIONAL SPEECH COMMUNICATION ASSOCIATION 2009, VOLS 1-5, 2009, : 1407 - 1410
  • [34] Speech Recognition Using HMM-CNN
    Santos, Lyndaines
    Moreira, Nicolas de Araujo
    Sampaio, Robson
    Lima, Raizielle
    Mattos Brito Oliveira, Francisco Carlos
    INFORMATION SYSTEMS AND TECHNOLOGIES, VOL 1, WORLDCIST 2023, 2024, 799 : 528 - 537
  • [35] DELETED SMOOTHING OF HMM PARAMETERS IN SPEECH RECOGNITION
    KIM, NS
    UN, CK
    ELECTRONICS LETTERS, 1993, 29 (09) : 735 - 736
  • [36] A SPECTRAL FEATURE PROCESS FOR SPEECH RECOGNITION USING HMM WITH MFCC APPROACH
    Shigli, Ashok
    Patel, Ibrahim
    Rao, Srinivas
    2012 NATIONAL CONFERENCE ON COMPUTING AND COMMUNICATION SYSTEMS (NCCCS), 2012, : 219 - 224
  • [37] Word Embeddings for Speech Recognition
    Bengio, Samy
    Heigold, Georg
    15TH ANNUAL CONFERENCE OF THE INTERNATIONAL SPEECH COMMUNICATION ASSOCIATION (INTERSPEECH 2014), VOLS 1-4, 2014, : 1053 - 1057
  • [38] A FPGA-based HMM for a discrete arabic speech recognition system
    Elmisery, FA
    Khalil, AH
    Salama, AE
    Hammed, HF
    ICM 2003: PROCEEDINGS OF THE 15TH INTERNATIONAL CONFERENCE ON MICROELECTRONICS, 2003, : 322 - 325
  • [39] A novel HMM model adaptation and compensation method for robust speech recognition
    Ning, GX
    Wei, G
    INTERNATIONAL SYMPOSIUM ON COMMUNICATIONS AND INFORMATION TECHNOLOGIES 2005, VOLS 1 AND 2, PROCEEDINGS, 2005, : 274 - 277
  • [40] Application of an Isolated Word Speech Recognition System in the Field of Mental Health Consultation: Development and Usability Study
    Fu, Weifeng
    JMIR MEDICAL INFORMATICS, 2020, 8 (06)
12345