The New Technique Enhancing of Automatic Speech Recognition System for ODIA Language using HTK Based On Hidden Markov Model (HMM)
Priyabrata Sahu1, Sunil Ku Panigrahy2, Umakant Bhaskar Gohatre3

1Priyabrata Sahu *, Assistant Professor in the Department of Computer Science Engineering and Applications, Indira Gandhi Institute of Technology, Odisha, India.
2Sunil Ku Panigrahi, Research Scholar, Department of computer Engineering, KIIT University, Odisha, India.
3Umakant Bhaskar Gohatre, Assistant Professor, Department of Engineering, University of Mumbai, Mumbai, India.

Manuscript received on April 30, 2020. | Revised Manuscript received on May 06, 2020. | Manuscript published on May 30, 2020. | PP: 2182-2187 | Volume-9 Issue-1, May 2020. | Retrieval Number: A2817059120/2020©BEIESP | DOI: 10.35940/ijrte.A2817.059120
Open Access | Ethics and Policies | Cite | Mendeley
© The Authors. Blue Eyes Intelligence Engineering and Sciences Publication (BEIESP). This is an open access article under the CC-BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/)

Abstract: The purpose of this paper is to address the application to an Indian Regional Language, ODIA of a single word Automatically Speech Recognition System (ASRS). The toolkit is based on Hidden Markov Model (HMM). The details was obtained from 8 ODIA Language speakers. The program is then qualified for 205 different terms in ODIA. Samples from six separate speakers have again been obtained. This is then evaluated in real time. A GUI has been created to enhance the system’s interactivity. We used and introduced the test framework for development of the GUI JAVA application. A comprehensive model of an ASR framework was developed to explain each HTK resource using HTK library modules and software. The findings of the experiment indicate that the overall machine efficiency is 93.45%. 
Keywords: HMM, HTK, P-ASR, Automatic Speech Recognition system.
Scope of the Article: Recognition System