Language Identification Performance Evaluation Using Spectral Processing

7 Pages Posted: 30 Nov 2020

See all articles by Deepti Deshwal

Deepti Deshwal

IFTM University

Pardeep Sangwan

Maharaja Surajmal Institute of Technology

Divya Kumar

IFTM University

Date Written: November 21, 2020

Abstract

Language Identification (LID) is a method by which the language is identified from an utterance communicated by an unidentified person. Present work develops an effective baseline method with the Gaussian Mixture Model (GMM) and Mel Frequency Cepstral Coefficients (MFCC) for language identification and the performance of the LID system is evaluated for an unknown speaker. The LID framework is designed utilizing a user-defined database in 4 different languages Tamil, Malayalam, Hindi, and English. This work is based on some optimization approaches such as Minimum Mean Square Error (MMSE) and Spectral Subtraction (SS) to improve the LID performance with background noise. Moreover, the LID system performance is also investigated by changing the size of data used to train the system, duration of test data as well as the amount of noise.

Keywords: Gaussian Mixture Modelling, Language identification, Mel frequency cepstral coefficients, Minimum Mean Square Error, Spectral Subtraction

Suggested Citation

Deshwal, Deepti and Sangwan, Pardeep and Kumar, Divya, Language Identification Performance Evaluation Using Spectral Processing (November 21, 2020). Proceedings of the 2nd International Conference on IoT, Social, Mobile, Analytics & Cloud in Computational Vision & Bio-Engineering (ISMAC-CVB 2020), Available at SSRN: https://ssrn.com/abstract=3734808 or http://dx.doi.org/10.2139/ssrn.3734808

Deepti Deshwal (Contact Author)

IFTM University ( email )

Lodhipur Rajput
Delhi Road
Moradabad, 244001
India

Pardeep Sangwan

Maharaja Surajmal Institute of Technology ( email )

C-4, Lal Sain Mandir Marg, Janak Puri
New Delhi, Delhi 110058
India

Divya Kumar

IFTM University ( email )

Lodhipur Rajput
Delhi Road
Moradabad, 244001
India

Do you have a job opening that you would like to promote on SSRN?

Paper statistics

Downloads
31
Abstract Views
175
PlumX Metrics