Dissertation > Excellent graduate degree dissertation topics show

Study of Speech Recognition System for Mandarin Digit Based on HMM

Author: HouZhouGuo
Tutor: QianShengYou
School: Hunan Normal University
Course: Circuits and Systems
Keywords: Speech recognition Linear Predictive Cepstral Mel Frequency Cepstral Dynamic Time Warping Hidden Markov Models
CLC: TN912.34
Type: Master's thesis
Year: 2006
Downloads: 349
Quote: 15
Read: Download Dissertation

Abstract


The voice is an important tool of communication between man and machine , voice recognition technology is to allow the machine to understand the human voice and the implementation of related action , with a wide range of application background . Although a lot of research in this area , but there are still many issues to be explored further. The acoustic models and voice recognition theory is to build the basis of a speech recognition system . This paper first introduces the voice acoustic model structure , a detailed analysis of each processing step and then follow the voice recognition system . Improved spectral entropy algorithm in the voice signal endpoint detection , experiments show that using the method of endpoint detection and feature extraction parameters to improve the robustness of the speech recognition system . Selection of characteristic parameters have a great impact on the entire real-time speech recognition system , robustness . Analysis of the short-term time-domain characteristics of the speech signal and spectrogram elaborate linear prediction coefficients ( LPC ??) , Linear Predictive Cepstral (LPCC) and Mel Frequency Cepstral Coefficients (MFCC) feature parameters extraction method , and its distortion measure . This paper discusses the dynamic time warping theory and hidden Markov model principle , speech recognition system using MATLAB programming language . Specific and non- specific recognition and selection of characteristic parameters of the recognition rate compared with the the DTW theory isolated word speech recognition . In addition , the actual building of the whole system of small isolated word speech recognition based on the HMM model non- specific people , the system can choose different characteristic parameters , has better robustness . Recognition experiments with the system of \about 10% higher .

Related Dissertations

  1. Technology of Tibetan Speech Recognition Based on Fast Walsh Transform,TN912.34
  2. The Research and Implement of Isolated Word Speech Recognition Algorithm Based on DTW Model,TN912.34
  3. Research on the Automatic Classification of Cough,TN912.34
  4. Multiple ANN/HMM Hybrid Used in Speech Recognition,TN912.34
  5. A Study on Effectiveness of an ASR-based Online Software on Chinese College Students’ Phoneme Learning,H319
  6. A Study of Speech Enhancement and Recognition Based on Microphone Array Processing,TN912.35
  7. Design and Realization of the Speech Recognition System Based on MCU,TN912.34
  8. RBF neural network optimized for speech recognition research,TN912.34
  9. Research and Emplementation of Embedded GUI Based on Qt/Embedded and Qtopia,TP368.12
  10. Study on Engine Monitoring and Fault Diagnosis Methods Based on Exhaust Pressure Wave Analysis,U472.43
  11. K neighbors phased approach in the wireless LAN intrusion detection research,TN925.93
  12. Gait acceleration signal based authentication method,TN911.7
  13. Confusion network -based classification of speech document theme,TN912.3
  14. The Chinese vowel length adjustment based speech recognition,TN912.34
  15. HMM and GMM based on natural and artificial blasting earthquake Recognition Algorithm,TN911.7
  16. Extended information extraction based on semantic technology research,TP391.1
  17. Mobile robot voice recognition control simulation system design and implementation,TN912.34
  18. Topic Classification of Speech Documents Based on the Word Fragment Network,TN912.3
  19. Statistical Image Modeling and Image Segmentation,TP391.41
  20. Moving target trajectory based recognition system for human-computer interaction,TP391.41
  21. Voice-based control of the electric car design,TP273

CLC: > Industrial Technology > Radio electronics, telecommunications technology > Communicate > Electro-acoustic technology and speech signal processing > Speech Signal Processing > Speech Recognition and equipment
© 2012 www.DissertationTopic.Net  Mobile