Audio-Visual Based Multi-Sample Fusion to Enhance Correlation Filters Speaker Verification System
Journal Title: International Journal on Computer Science and Engineering - Year 2010, Vol 2, Issue 4
Abstract
In this study, we propose a novel approach for speaker verification system that uses a spectrogram image as features and Unconstrained Minimum Average Correlation Energy (UMACE) filters as classifiers. Since speech signal is a ehavioral signal, the speech data has a tendency not to consistently reproduce due to the change of speaking rates, health, emotional conditions, temperature and humidity. In rder to overcome this problem, a modification of UMACE filters architecture is proposed by executing a multi-sample fusion using speech and lipreading data. So as to evaluate the utstanding fusion scheme, five multisample fusion strategies, i.e. maximum, minimum, median, average and majority vote are first experimented using the speech signal data. Afterward, the performance of the audiovisual system using the enhanced UMACE filters is then tested. Here, lipreading data is combined to the audio samples pool and the outstanding fusion scheme that found in prior experiment is used as multi-sample fusion scheme. The Digit Database had been used for performance evaluation and the performance up to 99.64% is achieved by using the enhanced UMACE filters for the speech only system which is 6.89% improvement compared with the base line pproach. Subsequently, the implementation of the audio-visual system is observed to be significant in order to broaden the SR score interval between the authentic and imposter data as well as to further improve the performance of audio only system that offer toward a robust verification system.
Authors and Affiliations
Dzati Athiar Ramli , Salina Abdul Samad , Aini Hussain
Keywords
Related Articles
- EP ID EP150201
- DOI -
- Views 126
- Downloads 0
An Integer Programming-based Local Search for Large-scale Maximal Covering Problems
Maximal covering problem (MCP) is classified as a linear integer optimization problem which can be effectively solved by integer programming technique. However, as the problem size grows, integer programming requires exc...
AGE CLASSIFICATIONS BASED ON SECOND ORDER IMAGE COMPRESSED AND FUZZY REDUCED GREY LEVEL (SICFRG) MODEL
One of the most fundamental issues in image classification and recognition are how to characterize images using derived features. Many texture classification and recognition problems in the literature usually require the...
Anti-Synchronization of the Hyperchaotic Lorenz Systems by Sliding Mode Control
This paper investigates the problem of global chaos anti-synchronization of identical hyperchaotic Lorenz systems (Jia, 2007) by sliding mode control. The stability results derived in this paper for the anti-synchronizat...
A System Based Approach to Efficiently Implement Color Differentiation Mechanism with Respect to Automobiles
Artificial Intelligence is the intelligence of machines focusing on creating machines that can engage on behaviors that human considers Intelligent Expert Systems are the programs based on their theory and methods of AI....
Enhancing Modularity in Aspect-Oriented Software Systems-An Emperical Study
Aspect-oriented programming (AOP) is rapidly gaining popularity among research and industry as a methodology that complements and extends the object-oriented paradigm.AOP promises to localize the concerns that inherently...