Gender Effect Canonicalization for Bangla ASR
Journal Title: International Journal of Advanced Computer Science & Applications - Year 2012, Vol 3, Issue 11
Abstract
This paper presents a Bangla (widely used as Bengali) automatic speech recognition system (ASR) by suppressing gender effects. Gender characteristic plays an important role on the performance of ASR. If there is a suppression process that represses the decrease of differences in acoustic-likelihood among categories resulted from gender factors, a robust ASR system can be realized. In the proposed method, we have designed a new ASR incorporating the Local Features (LFs) instead of standard mel frequency cepstral coefficients (MFCCs) as an acoustic feature for Bangla by suppressing the gender effects, which embeds three HMM-based classifiers for corresponding male, female and geneder-independent (GI) characteristics. In the experiments on Bangla speech database prepared by us, the proposed system has achieved a significant improvement of word correct rates (WCRs), word accuracies (WAs) and sentence correct rates (SCRs) in comparison with the method that incorporates Standard MFCCs.
Authors and Affiliations
B. K. M Mizanur Rahman , Bulbul Ahamed , Md. Asfak-Ur-Rahman , Khaled Mahmud , Mohammad Nurul Huda
Keywords
Related Articles
- EP ID EP114528
- DOI 10.14569/IJACSA.2012.031115
- Views 98
- Downloads 0
Identifying Green Services using GSLA Model for Achieving Sustainability in Industries
Green SLA (GSLA) is a formal agreement between service providers/vendors and users/customers incorporating all the traditional/basic commitments (Basic SLAs) as well as incorporating Ecological, Economical, and Ethical (...
Optimized Quality Model for Agile Development: Extreme Programming (XP) as a Case Scenario
The attributes of quality are that it is complex taxonomy, it cannot be weighted or measured but can be felt, discussed and judged. Early assessment and verification of functional attributes (requirements) are supported...
From Emotion Recognition to Website Customizations
A computer vision system that recognizes the emotions of a website’s user and customizes the context and the presentation of this website accordingly is presented herein. A logistic regression classifiers is trained over...
A Novel Permutation Based Approach for Effective and Efficient Representation of Face Images under Varying Illuminations
Paramount importance for an automated face recognition system is the ability to enhance discriminatory power with a low-dimensional feature representation. Keeping this as a focal point, we present a novel approach for f...
The Degree to which Private Education Students at Princess Nourah Bint Abdulrahman University have Access to Soft Skills from their Point of View and Educational Body
The study aimed at identifying the degree of ownership of special education students in the Department of Special Education, Faculty of Education, Princess Nourah University for soft skills from their point of view and t...