Please use this identifier to cite or link to this item: http://hdl.handle.net/123456789/1722
Full metadata record
DC FieldValueLanguage
dc.contributor.authorDeka, Aniruddha-
dc.contributor.authorDeka, Dr. Manoj Kr.-
dc.date.accessioned2018-08-02T06:25:31Z-
dc.date.available2018-08-02T06:25:31Z-
dc.date.issued2015-03-
dc.identifier.issn2349-5162-
dc.identifier.urihttp://hdl.handle.net/123456789/1722-
dc.description.abstractBodo language is a tonal as well as Semitic language belonging to North-East of India and has a limited number of speakers in Assam. It comes under the Assam-Burmese group of languages. It is said to have branched off from the Tibeto-Burman family of languages and is spoken by Bodo people of Nepal and Bangladesh, apart from North-eastern India. Infect, the Bodo language happens to be amongst the official languages of the Indian state of Assam and is one of the 22 languages recognized by the eighth schedule of the Indian constitution. But in the field of speech processing a limited number of work has been done till date. Some work has been done in the field of isolated word. Connected word recognition is one area where no work has been done so far for Bodo language. In this paper, an effort has been made to build automatic speech recognizer to recognize Bodo digit from 0-9 by using acoustic modeling approach on HTK 3. 4. 1 speech engine. We have collected Bodo digit from 10 different speakers with all most all possible variation. In order to achieve all the possible tones for all the digits a permutation has been performed with the digits from 0 to 9 and then they are recorded. Overall recognition accuracy has been found to be 82. 12% at connected word modelen_US
dc.language.isoenen_US
dc.subjectBodo Digiten_US
dc.subjectIPA Symbolen_US
dc.subjectHMMen_US
dc.subjectHTKen_US
dc.titleConnected Digit Recognition System in Bodo Languageen_US
dc.typeArticleen_US
Appears in Collections:Mr. Aniruddha Deka

Files in This Item:
File Description SizeFormat 
JETIR1503049.pdf513.69 kBAdobe PDFView/Open


Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.