Incorporating finer acoustic phonetic features in lexicon for Hindi language speech recognition. Issue 8 (17th November 2019)

• Record Type: Journal Article
• Authors: Patil, Atul; More, Prashant; Sasikumar, M.
• Is Part Of: Journal of information & optimization sciences. Volume 40:Issue 8(2019)
• Page Start: 1731
• Page End: 1739
• Publication Date: 2019-11-17
• Subjects: Computer Science; Automatic Speech Recognition -- ASR -- Lexicon development -- Indian Language Speech Recognition; Electronic data processing -- Periodicals; Information science -- Periodicals; Mathematical optimization -- Periodicals; 519.6
• DOI: 10.1080/02522667.2019.1703266
• Languages: English
• ISSNs: 0252-2667

Abstract

Abstract: Lexicon, used in speech recognition, is the list of words and corresponding pronunciation using a standard acoustic notation. Currently ILSL12 phone-set is widely used for Indian Language speech recognition. However, it has limitations to represent features of the speech like, voicing, fricatives, etc. that are found in Indian Languages. This paper addresses this limitation by considering the voiced and unvoiced features for Hindi speech recognition by incorporating finer representations at the time of lexicon expansion. The approach is tested for Hindi word recognition and has shown significant improvement in WER.