Text independent language recognition system for indic languages with new features

摘要

Spoken Language Identification is a task of identifying the language of an unknown utterance of speech. This paper describes a text independent language identification system using vector quantization with new features derived from MFCC feature of speech signal with a common code book. In this work, MFCC feature vectors of speech signal are transformed into new feature vectors. This LID approach includes generation of a common codebook using vector quantization with new feature set, one for each language. The experiments are carried out on Indian languages consists of six languages namely Tamil, Hindi, Tamil, Marathi, Malayalam and Kannada.