Efficient Speech Recognition System for Isolated Digits
Journal Title: International Journal of Computer Science & Engineering Technology - Year 2013, Vol 4, Issue 3
Abstract
In this paper, an efficient speech recognition system is proposed for speaker-independent isolated digits (0 to 9). Using the Weighted MFCC (WMFCC), low computational overhead is achieved since only 13 weighted MFCC coefficients are used. In order to capture the trends of the extracted features, the local and global features are computed using the Improved Features for Dynamic Time Warping (IFDTW) algorithm. In this work, we propose to reduce the time complexity of the recognition system by time-scale modification using a SOLA-based technique and also by using a faster implementation of IFDTW. The experiments based on TI-Digits corpus demonstrate the effectiveness of proposed system giving higher recognition accuracy of 99.16% and performing about 22 times faster than conventional techniques.
Authors and Affiliations
Santosh V. Chapaneri , Dr. Deepak J. Jayaswal
Keywords
Related Articles
- EP ID EP120197
- DOI -
- Views 112
- Downloads 0
Implementation of Vehicle License Plate Recognition Using Canny Edge Detection
Vehicle number plate identification based on digital image processing is a vital area of research ongoing in present scenario. It offers various applications expanding through many fields. This technique may be used for...
An Overview of Energy Consumption Techniques in Wireless Sensor Networks
A Wireless Sensor Network (WSN) could be a wireless network consisting of spatially distributed autonomous devices that use sensors for watching and recording the physical conditions of the surroundings and organizing th...
A Scenario of Brain Computer Interaction with Different types of Face Recognition Techniques
Brain computer interaction technology has been developed to record the brain signal activity through various methods like Digital Electroencephalography, EMG , FMRI and face recognition. These are the advanced technology...
AN ANALYSIS OF EXISTING ARTIFICIAL STOCK MARKET MODELS FOR REPRESENTING BOMBAY STOCK EXCHANGE (BSE)
The goal of agent-based modeling of stock markets is to enrich our understanding of fundamental processes that appear in a market. Artificial stock markets are models of financial markets used to study and understand mar...
Recent Developments in Superresolution
Super resolution imaging refers to inferring the missing high resolution image from low resolution image(s). Super resolution methods are generally classified into reconstruction based and learning based methods. This pa...