Performance Evaluation of Loss Functions for Margin Based Robust Speech Recognition
Journal Title: International Journal of Advanced Computer Science & Applications - Year 2016, Vol 7, Issue 2
Abstract
Margin-based model estimation methods are applied for speech recognition to enhance the generalization capability of acoustic model by increasing the margin. An important aspects of margin based acoustic model for parameter estimation is that, the acoustic models are derived from soft margin concept and hinge loss function used in SVM as loss function to attained enhanced speech recognition performance. In this study, performance evaluation of loss functions (Logistic, Savage, Sigmoid) have been computed in the presence of white noise, pink noise, and brown noise with and without SVM classifiers to analyze the impact of noise on loss functions in comparison with hinge loss function used in SVM for parameter estimation in margin based acoustic model. Experimental results show that hinge loss function in the presence of pink noise and white noise have significant effects on isolated digits (0-9) in both pre-conditioned and recorded data samples in comparison with brown noise. Whereas hinge loss functions show serious anomalies with savage loss and sigmoid loss in term of performance and sigmoid loss function provides exceptionally good results in term of percentage error for all prescribed conditions.
Authors and Affiliations
Syed Ali, Maria Andleeb, Raheela Asif, Danish-ur-Rehman
Keywords
Related Articles
- EP ID EP159283
- DOI 10.14569/IJACSA.2016.070249
- Views 120
- Downloads 0
Socialization of Information Technology Utilization and Knowledge of Information System Effectiveness at Hospital Nurses in Medan, North Sumatra
Background of this research is the globalization and development of science, especially in the field of information and communication technology and communication that has influenced and has implications for changes and...
Experimental Evaluation of the Virtual Environment Efficiency for Distributed Software Development
At every software design stage nowadays, there is an acute need to solve the problem of effective choice of libraries, development technologies, data exchange formats, virtual environment systems, characteristics of virt...
Image Contrast Enhancement by Scaling Reconstructed Approximation Coefficients using SVD Combined Masking Technique
The proposed method addresses the general issues of image contrast enhancement. The input image is enhanced by incorporating discrete wavelet transform, singular value decomposition, standard intensity deviation based cl...
Arabic Sentiment Analysis: A Survey
Most social media commentary in the Arabic language space is made using unstructured non-grammatical slang Arabic language, presenting complex challenges for sentiment analysis and opinion extraction of online commentary...
Fall Monitoring Device for Old People based on Tri-Axial Accelerometer
To be able to timely and effective judgment of the elderly fall, a fall monitoring device based on tri-axis accelerometer for elderly is designed. The device collects acceleration and the angle between elderly and horizo...