Dissertation > Excellent graduate degree dissertation topics show

The Study of Car Speech Recognition Based on the Specific Human and Small Vocabulary

Author: DongXiangLin
Tutor: HuangTao
School: Wuhan University of Technology
Course: Communication and Information System
Keywords: Speech recognition DTW model Car voice Endpoint detection
CLC: TN912.34
Type: Master's thesis
Year: 2010
Downloads: 210
Quote: 1
Read: Download Dissertation

Abstract


In information technology, human-computer interface, speech recognition and processing technology to get widespread attention in its electronic products to make people's lives more exciting. People will be able to control system devices via voice commands, let the appropriate action to respond to voice commands. This has the voice recognition system has a very important value in the Internet, communications, military, national defense. The voice recognition technology used in car platform, which enables the driver of the car even more flexible, simple, more safety and comfort. DTW model and algorithm of this paper is based on a specific small vocabulary speech recognition technology. The basic method of speech recognition and speech recognition algorithm based on the traditional DTW algorithm improvement and optimization. In this paper, the variable window length, and the double-threshold method of combining voice endpoint detection. During optimal path selection, taken relaxation starting point and the end of a way to select the optimal matching path. The DTW algorithm recognition results improved significantly optimize the recognition result of the traditional DTW algorithm through MATLAB simulation results can be seen. The full text The first is the basic principle of the speech recognition technology has made the introduction and analysis. For specific isolated word vocabulary speech recognition system, this choice of DTW algorithm for speech recognition. Determine the selection of the DTW algorithm DTW algorithm improvement and optimization. Improved DTW algorithm with traditional DTW algorithm and compared by the comparison of the simulation results, we can see that the optimized algorithm is superior to the traditional algorithm. During endpoint detection process, the article first after the processing of the sub-frame, the speech signal is divided as a silent segment, the transition section and the speech segment. Different window length to be processed and then were taken on mute stage, transitional stage, voice segment. This selection of a longer window length in the silent segment for processing, the voice transition section to take a smaller window length and the frame shift in the speech segment, we take a conventional window, so that this will not affect the processing speed of the speech recognition system, they can be more accurate endpoint detection purposes. During the long process Variable This article also incorporates dual-threshold endpoint detection methods to voice signal endpoint detection. Specific DTW algorithm implemented process, the use of dynamic of tacticity and relaxation endpoint method to select the optimal matching path. Specific hardware implementation, this article uses minimum system with the highest cost-effective solutions to implement voice recognition function. Speech recognition module completely homemade program, and in control of the car, this paper divided band transmit waveform to control the car in response to different actions. For this voice recognition system, this paper presents the areas for improvement. Finally, we summarize the full text of the work done, and the future of speech recognition prospect.

Related Dissertations

  1. Multiple ANN/HMM Hybrid Used in Speech Recognition,TN912.34
  2. The Design of a DSP-Based Robot Speech Command Recgnition System,TN912.34
  3. The Design and Research of Health Management Based on Smartphone Environment,TN929.53
  4. Mobile audio and video interactive platform of business execution,TN915.09
  5. Power spectrum estimation in the broadband ADCP Signal Detection Research and Application,TN911.23
  6. Telephone-based channel voiceprint recognition algorithm,TN912.34
  7. Research on Hmm-based Speech Recognition System of the Robot,TN912.34
  8. MFCC -based speech recognition system to improve research and design,TN912.34
  9. Robot Control System Simulation,TP242
  10. Research and Implemenation of Voice Intelligent plantform Based on VoiceXML,TP311.52
  11. Topic Classification of Speech Documents Based on the Word Fragment Network,TN912.3
  12. Study on Hybrid Model of Speech Recognition Based on HMM and PNN,TN912.34
  13. Mobile robot voice recognition control simulation system design and implementation,TN912.34
  14. Research on Endpoint Detection Algorithm and Implementation in Hardware,TN912.34
  15. Research on DBN-Based Continuous Speech Recognition,TN912.34
  16. STRAIGHT spectrum - based speech recognition algorithm research,TN912.34
  17. Endpoint Detection of Speech Signal Based on Empirical Mode Decomposition,TN912.3
  18. Research on the Key Technologies fo Speech Recognition for Robot Communication,TN912.34
  19. The LVCSR system based on adaptive methods of semi-supervised learning,TN912.34
  20. Parallel Optimization Method in Language Model for Mandarin Speech Recognition,TN912.34
  21. Research of Segmentation Based Chinese Continuous Speech Recognition Technology,TN912.34

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