Login | Signup now       

Guest

 Click to see how  

HOME | VIDEOS | DOCUMENTS | COLLECTIONS | UPLOAD | BROADCAST | MY ACCOUNT | FEEDBACK | ABOUT

 

Share  Title: Evaluation of Phase & Magnitude based Features for Speaker I...

 

 
Abstract

 Download Papers  Read full document

 
 
 
iConnect (Beta)   |   Sponsor  |   Comment   |    Report    Like    
 
 

Given a speech signal, there are two most important information that can be extracted from it. One being the linguistic information (about what is being said) and other being the speaker specific information (about who is speaking). This report is about the task of speaker recognition where the goal is to determine the speaker identity, from a group of known speaker, which closely matches with input sample. This problem become even more tough when there is limited amount of test and train data, a mismatch between the surrounding conditions while recording the test and train data, or in noisy environments. In this thesis, we consider the problem of speaker identification in noisy and bandlimited telephonic environments using the Gaussian Mixture Model approach combined with sub-band based feature extraction. We implement a sub-band based Posteriori Union Model described by Reynolds [1]. Then, we extend sub-band based approach to combine the phase based feature ModGDF and Magnitude based feature MFCC using several feature recombination techniques described in this thesis. These sub-band based feature recombination methods gives 46% identification accuracy, in best case, on NTIMIT database with little or no increase in computation.

 
 
Download Papers  Read full document
 
 
 
Most Viewed This Month
Supersonic Nozzle design by Method of Characteristics
A report discussing critical factors affecting Project performance
JPEG Progressive mode vs Sequential mode
Business Plan for Online Money Transfer
Rayleigh criteria and Combustion Instability
Adaptive Equalization Techniques using Least Mean Square (LMS) algorithm
Girlkultur Mass Culture in Weimar Germany
Adaptive Equalization Techniques using Recursive Least Square (RLS) algorithm
Combustion Instability in Liquid Rocket Engines
Laminar Separation Bubble
A Technical Essay on the Gyroplane
Experimental Study of Compressible Pipe Flow with Friction and Heat Addition
The Role of Density Gradient in Liquid Rocket Engine Combustion Instability
NOTAR Helicopter
Blind De-convolution of BPSK and QPSK modulated signal using Constant Modulus Algorithm (CMA).
 
Related Documents
Evaluation of Phase & Magnitude based Features for Speaker Identification
Evaluation of phase and magnitude based features for speaker identification
Evaluation of Feature Recombination Techniques for Speaker Identification
Combustion Instability in Liquid Rocket Engines
Gender Classification Using Support Vector Machines (SVMs)
A Machine Learning approach to localization in Wireless Sensor Networks
Application of Multiresolution Analysis in Renormalization Group Calculations
Hydrogen sensors based on titania nano-tubes
The Role of Density Gradient in Liquid Rocket Engine Combustion Instability
A Comparative Study of Different Face Recognition Algorithms
Laminar Separation Bubble
Integrated Approach to Value Management and Sustainable Construction in Indian Construction Industry
Computational Support for the Collaborative Design, Routing, and Manipulation of Cable Harnesses
Temperature measurements during Reentry
Hints for understanding low speed wind tunnels
 
Member Documents
Adaptive Equalization Techniques using Least Mean Square (LMS) algorithm
Adaptive Equalization Techniques using Recursive Least Square (RLS) algorithm
Blind De-convolution of BPSK and QPSK modulated signal using Constant Modulus Algorithm (CMA).
Digital Image Processing
JPEG Progressive mode vs Sequential mode
Evaluation of Phase & Magnitude based Features for Speaker Identification
A Comparative Study of Different Face Recognition Algorithms
Gender Classification Using Support Vector Machines (SVMs)
Evaluation of phase and magnitude based features for speaker identification
Evaluation of Feature Recombination Techniques for Speaker Identification
A Machine Learning approach to localization in Wireless Sensor Networks
Complex Classification Using Advanced Machine Learning Techniques
 

Keywords

speaker  identification  mfcc  modified  group  delay  
sub-band  recombination  missing  feature  theory  
  
 
 
 About This Document
 
 Subject Computer Science
 Category Term Paper
 Views 3350
 Downloads 282
 Added 25-03-10
 Contributor   ravigarg
 Add to Favourites
 Report Abuse
 
 Related Videos
 See More

 RunTime  00:00:02
 Uploaded  21-11-07
 Views  3391
   
 Moebius Transformations R...

 RunTime  moderate
 Uploaded  24-04-11
 Views  3738
   
 Patent Application

 RunTime  00:00:02
 Uploaded  31-10-07
 Views  3323
   
 Cellular Division

 RunTime  00:10:21
 Uploaded  28-11-07
 Views  4065
   
 Chaos and Fractals in Sim...

 RunTime  00:09:24
 Uploaded  28-11-07
 Views  4110
   
 Chaos and Fractals in Sim...

 RunTime  00:10:36
 Uploaded  28-11-07
 Views  3855
   
 Chaos and Fractals in Sim...

 RunTime  00:00:08
 Uploaded  22-11-08
 Views  3420
   
 Sleep Apnea

 RunTime  00:17:23
 Uploaded  20-12-07
 Views  3312
   
 Geometry + 4-Potentials =...

 RunTime  moderate
 Uploaded  02-09-11
 Views  2191
   
 Improving the safety of a...

 RunTime  00:04:44
 Uploaded  26-08-09
 Views  3393
   
 Beats

 RunTime  00:06:50
 Uploaded  10-01-08
 Views  4644
   
 Lisa Randall Lecture

 RunTime  00:04:21
 Uploaded  14-12-07
 Views  3899
   
 GE90-115B Gas Turbine Jet...

 RunTime  00:08:30
 Uploaded  26-12-07
 Views  7168
   
 How to Design a Boat Hull...

 RunTime  00:10:31
 Uploaded  26-08-09
 Views  3330
   
 Interference

 RunTime  00:09:49
 Uploaded  24-12-07
 Views  3729
   
 Oscilloscope demo

 RunTime  00:09:58
 Uploaded  30-04-08
 Views  3810
   
 The Elegant Universe : Ca...

 RunTime  00:03:07
 Uploaded  28-12-07
 Views  3451
   
 Radiometric Dating Revisi...

 RunTime  moderate
 Uploaded  10-09-10
 Views  4385
   
 Whales originated from aq...

 

 

Comments | Queries | Clarifications