SPEAKER IDENTIFICATION USING 2-D DCT, WALSH AND HAAR ON FULL AND BLOCK SPECTROGRAM
Journal Title: International Journal on Computer Science and Engineering - Year 2010, Vol 2, Issue 5
Abstract
This paper aims to provide different approaches to text dependent speaker identification using DCT, Walsh and Haar transform along with use of spectrograms. Spectrograms obtained from speech samples are used as image database for the study undertaken. This image database is then bjected to various transforms. Using Euclidean distance as measure of similarity, most appropriate speaker match is btained and is declared as identified speaker. Each transform is applied to spectrograms in two different ways: on full image and on mage blocks. In both the ways, effect of different umber of coefficients of transformed image is observed. Haar transform on full image reduces multiplications required by DCT and Walsh by 28 times whereas applying Haar transform on image blocks requires 18 times less mathematical omputations as compared to DCT and Walsh on image blocks. Transforms when applied to image blocks, yield better or equal identification rates with educed computational complexity.
Authors and Affiliations
Dr. H. B. Kekre , Dr. Tanuja K. Sarode , Shachi J. Natu , Prachi J. Natu
Keywords
Related Articles
- EP ID EP91894
- DOI -
- Views 117
- Downloads 0
A Genetic Algorithm Optimized Decision Tree- SVM based Stock Market Trend Prediction System
Prediction of stock market trends has been an area of great interest both to researchers attempting to uncover the information hidden in the stock market data and for those who wish to profit by trading stocks. The extre...
An Analysis on Preservation of Privacy in Data Mining
Privacy has become a key issue for progress in data mining. Maintaining the privacy of data mining has become ncreasingly popular because it allows sharing of privacy-sensitive data for analysis. So people are still rel...
Role Oriented Test Case Generation for Agent Based System
Agent Oriented Software Engineering (AOSE) is a rapidly developing area of research. Current research and development primarily focuses on the analysis, design and implementation of agent based software whereas testing i...
Intelligent Data Compression Approach in Multidimensional Data Warehouse
The problem with MOLAP is that large tables should be loaded in main memory, which can slow the system, even saturate the memory. In this work, we present a new compression method, called BTC, for multidimensional data w...
Multi Subgroup Data Compression Technique Using Switch Code
Data compression is the art of converting a data stream into a small in size data bits so that it can be easily travel a long istance without increasing load of its volume on a constant Bandwidth channel regardless of i...