In this paper, we introduce the first large vocabulary speech recognitionsystem (LVSR) for the Central Kurdish language, named Jira . The Kurdish language has several dialects and sub-dialects that results in many lexical variations . The system was trained with AsoSoft Speech-Office and Crowdsourcing and a combination of them. The best performance achieved by the SGMM acoustic model was 13.9% of the average word errorrate (on different document topics) and 4. 9% for the general topic. To setup therecognition engine, we used the Kaldi toolkit. A statistical tri-gram languagemodel that is extracted from the Asosoft text corpus is used in the system is used . To setup the system, we . used the . Kaldi Toolkit. to set the system’s recognition engine. To build the system. For more information, please visit www.kalditoolkit.com/kaldi.com.uk . Back to the page you came from: http://www.jaldi.uk/jira/kdal/kaldin.com/. Back to page you go to: www.jira@jira.com-jira-report.uk.com . Back page you went to: kaldin/kardin.org/kandalin-report .

Author(s) : Hadi Veisi, Hawre Hosseini, Mohammad Mohammadamini, Wirya Fathy, Aso Mahmudi

Links : PDF - Abstract

Code :
Coursera

Keywords : jira - system - speech - kaldi - www -

Leave a Reply

Your email address will not be published. Required fields are marked *