Dissertation > Excellent graduate degree dissertation topics show

Research on Whispered Speaker Identification in Channel Mismatch Conditions

Author: GuXiaoJiang
Tutor: ZhaoHeMing
School: Suzhou University
Course: Signal and Information Processing
Keywords: whispered speech speaker identification joint factor analysis hybrid compensation support vector machine
CLC: TN912.34
Type: Master's thesis
Year: 2011
Downloads: 29
Quote: 1
Read: Download Dissertation

Abstract


The whispered speech is acted as an auxiliary way of communication and it is widely used in human life at the same time, especially in the all kinds of identity recognition of finance area and justice area. Speaker usually can use whispered speech in order to keep information secret.So, the whispered speaker identification is also noticed as a new project. The whispered speech is often used in mobile phone environment, which is affected by channel distortion. The traditional model gets low recognition accuracy when the channel environment difference between training and testing is obvious. Therefore, a robust channel compensation algorithm must enhance the speaker recognition system. In order to solve this problem, the article’s work is as follows:1. Mix all the kinds of channel whispered speech to train a universal background model (UBM), then on this base, maximum a posteriori adaptation is adopted to train the speaker model. Compare this model with GMM, the experiment result proves that the UBM performs better than normal GMM.2. Joint factor analysis (JFA) is introduced in whispered speaker identification. According the speech database’s characteristic, decoupled estimation and omitting residual subspace are applied. In the specific identification process, the speaker factor from training utterance and channel factor from testing utterance are combined to fit the test channel dynamically. The experiment shows that improvement JFA achieves high recognition result. In addition, JFA is not ideal in the short-time identification. A new hybrid compensation method which keeps speaker factor in model domain and applies channel factor in feature domain is proposed. This method is to compensate each frame feature vector and more meticulous than JFA. The experiment shows 1s and 2s average identification rate separately improve 4.36% and 3.89% when HH channel is trained. In addition, EP channel separately improve 4.14% and 2.64%.3. According to support vector machine (SVM)’s discriminability, the speaker supervector is input into the SVM. But the system performance is not as good as UBM-MAP. Then the speaker factor vector is input into the SVM. Because the speaker factor has the property of low dimension and linear discriminant availability, it achieves excellent accuracy result. After that, three kinds of channel compensation technique are used to improve the system’s robustness further and obtain quite identification result compared to JFA.

Related Dissertations

  1. Research on Automatic Detection Algorithm for Substructure Distress of Highway Pavement Based on SVM,U418.6
  2. Compensation Methods of Different Speech Coding for Speaker Recognition,TN912.34
  3. Research on Autamatic Music Structrue Analysis,TN912.3
  4. Fault Diagnosis Method Based on Support Vector Machine,TP18
  5. Prediction of Binding Affinity of Human Transporter Associated with Antigen Processing,R392.1
  6. Unbalanced data set classification method and its application in the telecommunications industry,TP311.13
  7. Based on a sample study of the finite element mesh model optimization,TB114
  8. Research on Feature Selection for Gene Expression Data,Q78
  9. Based on SVM predictor Mine Gas Emission Study,TD712.5
  10. Research on Estimation and Prediction of Urban Link Travel Time Based on Low-Frequency Floationg Car Data,U491
  11. Requirement Analysis of System Ability Index of Weapon & Equipment System-of-Systems Based on SVM,E920
  12. The Study on Embedded Condition Detection and Fault Diagnosis Technology for Chassis of Self-propelled Gun,TJ818
  13. Multi Phase Intelligent Fuzzy Simulation Control Research and,U491.54
  14. Study on Short-Term Power Load Prediction Based on Support Vector Regression Modeling Method,TM715
  15. Appliciation of Support Vector Machines in Power System Load Forecasting,TM715
  16. Research on MPP of Partially Shaded PV System Based on SVR Algorithm,TM615
  17. Research on Bid Evaluation Method of Power Procurement Based on Support Vector Regression,F224
  18. Research on End-point Prediction Model of Electric Arc Furnace Based on SVM,TM924.4
  19. HVDC transmission system fault diagnosis method,TM721.1
  20. Short - term load forecasting , energy management system based on support vector machines,TM715
  21. Integrated Measurement for Density and Viscosity of Heavy Medium Suspension Based on Differential Pressure,TD94

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