Dissertation > Excellent graduate degree dissertation topics show

A Weighted Approach of Missing Data Technique Based on Sigmoid Function in Cepstral Domain

Author: YiPei
Tutor: ChenJinWen
School: Tsinghua University
Course: Mathematics
Keywords: speech recognition missing data technique reliability mismatch variance inflation factor
CLC: TN912.34
Type: Master's thesis
Year: 2010
Downloads: 50
Quote: 0
Read: Download Dissertation

Abstract


Recently the mainstream speech recognizer, based on Hidden Markov Model (HMM) and the Mel-frequency cepstral coefficients, has achieved good results in quiet environment. However, the performance deteriorates significantly in the presence of noise and other disturbance. As a result, many algorithms based on feature compensation or model adaptation, such as vector Taylor series, maximum linear likelihood regression, have been proposed to solve the problem.Missing data technique (MDT) has been proved a effective method in robust speech recognition. Conventionally, the identification of reliability is performed in spectral or log-spectral domain, and then the recognition is accomplished in log-spectral domain or cepstral domain followed by discrete cosine transformation. In order to apply MDT in cepstral domain, this paper presents a weighted average transformation to calculate the reliability of cepstral feature. The reliability is used as a weighting factor in the computation of Mahalanobis distance in HMM. Based on the additive noise model, the paper offers an origin about the weighted average transformation. It is deduced that the reliability enlarges the variance of HMM to reduce the mismatch between clean-training model and corrupted speech. This point is confirmed by the comparison of variances between in the clean-training model and in the multi-condition training model.To evaluate the performance of the propose algorithm, experiments are carried out on the Aurora2 speech database in the platform of HTK. The influence of parameters in the weighted MDT on the accuracy of recognition is analyzed carefully. and the algorithm is tested under the ideal condition and other two real condition. The result shows a distinct accuracy improvement compared to the baseline. Penalty factor and cooperation with other speech recognition methods are also tried in this paper. The main advantages of the weighted MDT are good interpretability, simple system implementation, low computation cost and easy to plug into other robust recognition algorithm. Finally, the research work is summarized.

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. Development and Preliminary Application of the Questionnaire on Emotional Labour of the Middle School Teacher,B841
  4. The Design and Research of Health Management Based on Smartphone Environment,TN929.53
  5. The Research on Reliability of Low-Voltage Power Line Communication,TM73
  6. The Research and Application of Software Reliability Test for User Right Management System,TP311.53
  7. Research on Subsea Pipeline Repair Coupling,TE973
  8. Mechanism of Preattentive Processing of Emotional Prosody Among Patients with Panic Disorder: A Mismatch Negativity Study,R749.7
  9. Molecular Diagnosis for Changes of Methylation of MLH1 Promoter of Arabidopsis Thaliana Induced by Cadmium Stress,X173
  10. Research and Application on Reliability Technology of Mine Hidrological Monitoring System,TD742.1
  11. A Mismatch Negativity Study of Visual Physical Stimulus in Patients of Panic Disorder as Measured by a Multi-feature Paradigm,R749.7
  12. A Study of Neuroelectrophysiology in Patients with Depression Disorder and Patients with Generalized Anxiety Disorder,R749.4
  13. A Research on Speech Recognition Technology,TN912.34
  14. Research on the Methods for Detecting Mismatch of Web Services Based on Bounded Model Checking,TP311.52
  15. Molecular Dynamics Simulations of Structural Relaxation of Two-dimensional Lattice Mismatch Epitaxial Aluminum Thin Film,O484.1
  16. Reliability Test and Analysis of ASON Network,TN929.1
  17. Study on Reliability for Machining Center,TG659
  18. An Efficient Multicast Scheme Based on Network Coding in Wireless Networks,TN929.5
  19. Reliability Robust Design of Impeller of Large Centrifugal Compressor,TH452
  20. Sensitivity Analysis on Fatigue Reliability of a Turbine Disk Using Finite Element Method,V232.3
  21. Fatigue Reliability Sensitivity Analysis and Reliability Robust Design of High Speed Train Wheel,U270.33

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