Dissertation > Excellent graduate degree dissertation topics show
Speaker Recognition Based on Swarm Intelligence and Blind Source Separation
Author: LiWeiJuan
Tutor: LiMing
School: Lanzhou University of Technology
Course: Communication and Information System
Keywords: Speaker Recognition Independent Component Analysis Particle swarm classifier Support Vector Machine
CLC: TN912.34
Type: Master's thesis
Year: 2010
Downloads: 44
Quote: 0
Read: Download Dissertation
Abstract
|
Speaker recognition in clean speech conditions have obtained good recognition rate , but because of the presence of noise , the voice signal is distorted , resulting in the training environment and test environment does not match , a serious impact on the recognition rate . Reflect the speaker's personality traits extracted from complex background noise speech parameters and design effective classification is the speaker recognition applied to the actual difficulty . The problems above , the departure From Noisy Speech parameter extraction and classification of these two aspects of the design , to put forward their own solutions , and experiments to test its feasibility . The main work includes : 1 for speaker identification in mixed noise environment affect the accuracy of speech recognition , consider the ICA algorithm is improved and applied to the speech signal de-noising process , blind source separation algorithm based on independent component analysis traditional implementations are based on the gradient , the convergence performance and solving performance depends on the choice of learning step , and the convergence is slow . In order to overcome these shortcomings, particle swarm algorithm to improve the independent component analysis , MFCC coefficients to enhance speech features . The test results show that the improved ICA algorithm can quickly and efficiently get the optimal solution of the BSS can effectively suppress ambient noise , thereby improving voice quality . For support vector machine training in the case of a large sample of slow shortcomings with Weight optimal location strategy to improve the quantum particle swarm optimization , through improved Michigan encoding scheme to encode the voice parameters , the tectonic classification rules fitness function , to achieve classifier design based on the weighted quantum particle swarm . Speaker recognition results show that the classifier has better anti-noise performance and higher recognition speed simulation results show that the improved WQPS-classifier and other classifiers , better anti-noise performance and identify speed.
|
Related Dissertations
- Research on Automatic Detection Algorithm for Substructure Distress of Highway Pavement Based on SVM,U418.6
- Research on Autamatic Music Structrue Analysis,TN912.3
- Research on Transductive Support Vector Machine and Its Application in Image Retrieval,TP391.41
- Fault Diagnosis Method Based on Support Vector Machine,TP18
- Process Support Vector Machine and Its Application to Satellite Thermal Equilibrium Temperature Prediction,TP183
- Research for Infrared Image Target Identification and Tracking Technology,TP391.41
- Study on the Road Condition Monitoring Based on Vehicular 3D Acceleration Sensor,TP274
- Research of Diagnosing Cucumber Diseases Based on Hyperspectral Imaging,S436.421
- The Research on Intrusion Detection System Based on Machine Learning,TP393.08
- Research on Improved K Neighbor Support Vector Machine Algorithm Faced Text Classification,TP391.1
- Research on Face Recognition Based on AdaBoost Algorithm,TP391.41
- Research on Feature Extraction, Selection and Classification Algorithms for Pulmonary CAD,TP391.41
- Research on Subimage Selection and Mathching Method for Synthetic Aperture Radar(SAR) Target Recognition,TN957.52
- Research of Facial Expression Recognition Algorithm,TP391.41
- Fundus Image Segmentation Based on SVM and Template Matching,TP391.41
- Research and Realization of License Plate Character Recognition Algorithm Based on SVM,TP391.41
- Cloud Classification Based on Geostationary Meteorological Satellite Imagery,TP391.41
- The Studies on Some Improvements of the GA and Their Applications in SVM,TP18
- The Research and Implementation of the Fault Management System for Triple Play,TP315
- Modulation Classification Algorithms of Digital Communication Signals,TN914.3
- Computer virus detection method,TP309.5
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
|