Dissertation > Excellent graduate degree dissertation topics show

Research on Chinese Syllable Evaluation Approach after Automatic Speech Recogniton

Author: WangXingJian
Tutor: LiLei
School: Beijing University of Posts and Telecommunications
Course: Signal and Information Processing
Keywords: Post-processing of the speech recognition Pinyin syntax Knowledge Base The anchor word Error correction candidate words
CLC: TN912.34
Type: Master's thesis
Year: 2010
Downloads: 44
Quote: 1
Read: Download Dissertation

Abstract


The rapid development of today 's speech recognition technology has begun to be applied in many fields , but many problems still exist . First , because of the background noise , speaking population of tone and expression are not standardized , and new words emerging , the correct rate of speech recognition in real environment has been less than satisfactory . Voice document retrieval began large-scale applications , many speech recognition after text requires extensive processing . Therefore , we take the post-processing of speech recognition results , in order to improve speech recognition effect . This thesis is mainly based on the robustness of the results of the Ministry of Education, Science and Technology Research Key Project \Complete a full speech recognition text processing system in the context of this project , information on the text-to-speech evaluation algorithm , in-depth study . The main contents of this article and work : 1 . Phonetic structures syntax Knowledge Base . Smooth introduction and comparative analysis of the phonetic N-gram model with data obtained more appropriate Chinese pinyin grammar model and data smoothing methods , and describes how to build a different syntax models and smoothing methods Pinyin Knowledge Base . 2. Anchor word speech recognition result positioning. The continuous phonetic N-gram model analysis and evaluation , summarize the corresponding law the auxiliary application Pinyin stability , and finally determine the anchor word . 3 error correction candidate recommended . This part describes Pinyin confusion rules corresponding similarity of phonetic and converted into a corresponding candidate word . Scoring candidate words , using phonetic similarity to measure the closeness of the original spelling . 4 . Algorithm is applied to the actual system . The practical application of the above algorithm to a complete speech recognition error detection and correction system , introduced its system framework and concrete realization explained in this work in which play the role of the comparative analysis of the effect of the actual application .

Related Dissertations

  1. Text Correction for ASR Result on the Platform of Intelligent Mobile Phone,TP391.1
  2. Rule-based and statistical - based combination the bilingual parallel to sentence the phrase alignment method,TP391.2
  3. Compensation Methods of Different Speech Coding for Speaker Recognition,TN912.34
  4. Research on Endpoint Detection Algorithm and Implementation in Hardware,TN912.34
  5. Parallel Optimization Method in Language Model for Mandarin Speech Recognition,TN912.34
  6. Research on Robust Music Identification in the Compressed Domain,TN912.34
  7. Design and Implementation of Voice Screening System for Public Security Based on Speaker Recognition,TN912.34
  8. RBF neural network optimized for speech recognition research,TN912.34
  9. Based on artificial neural network model voice conversion algorithm,TN912.34
  10. MIDI-based instrument control systems and automatic identification method notes,TN912.34
  11. Speech Input Solution Research Based on Cloud Computing,TN912.34
  12. Research on Speech Recognition System Based on MCU,TN912.34
  13. Research of Speech Recognition of Isolated Word Under Embedded Linux Operation,TN912.34
  14. A Study on the Characteristic and Intervention of the Continuous Phone-repeating Ability of the Hearing-impaired Children Aged from 3 to 5,TN912.34
  15. Hardware and Software Platform Design of Embedded Speech Recognition System,TN912.34
  16. Study of Music Recognition Algorithm Based on RBF Neural Network and HMM,TN912.34
  17. Speech Recognition Algorithm and Implementation,TN912.34
  18. The Study of Speaker Recognition,TN912.34
  19. A Robust Speech Recognition Approach Based on Sub-band Maximum Likelihood Linear Regression,TN912.34
  20. FPGA-based speech recognition system design and implementation,TN912.34
  21. Research on Mandarin Connected Digit Speech Recognition System Based on HTK,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