Speech recognizer for regional languages of Pakistan

Author

Ali, Syed Abid ; Ashraf, Bushra ; Owais, Hafiza Amna ; Saifuddin, Sakina

Author_Institution

Dept. of Comput. & Inf. Syst. Eng., N.E.D Univ. of Eng. & Technolgy, Pakistan

fYear

2014

fDate

22-24 April 2014

Firstpage

68

Lastpage

72

Abstract

This paper introduces the speech recognizer developed in the regional languages of Pakistan: Urdu, Punjabi, Sindhi and Pashto for performing three different supervised operations (Turn on/Turn off Door, Turn on/Turn off Light, Turn on/Turn off Fan) on real time input speech sample for single speakertaken from developed corpus consists of 24 samples (6 samples for each language) in four different languages of Pakistan. The objective of this work is to develop the speech recognizer in the regional languages of Pakistan for those people who do not understand and speak languages other than their native language. Experimental results of the proposed speech recognizer demonstrate expected outcome under minor or no noise interference.

Keywords

natural languages; speech recognition; Pakistan; Pashto; Punjabi; Sindhi; Urdu; corpus; native language; regional languages; speech recognizer; Feature extraction; Gaussian mixture model; Mel frequency cepstral coefficient; Speech; Speech processing; Speech recognition; Gaussian Mixture Model (GMM); Mel-Frequency Cepstral Coefficients (MFCC); Speech recognizer; VAD;

fLanguage

English

Publisher

ieee

Conference_Titel

Robotics and Emerging Allied Technologies in Engineering (iCREATE), 2014 International Conference on

Conference_Location

Islamabad

Print_ISBN

978-1-4799-5131-4

Type

conf

DOI

10.1109/iCREATE.2014.6828341

Filename

6828341