This paper describes the details of a language identification technique suitable for multilingual speech recognition systems. The present system is trained for two of the most widely spoken Indian languages, Tamil and Hindi. The technique could be extended for any number of languages without any substantial increase in the size of the multilingual speech recognition system.
Cite as: Kumar, C.S., Li, H. (2004) Language identification for multilingual speech recognition systems. Proc. 9th Conference on Speech and Computer (SPECOM 2004), 343-347
@inproceedings{kumar04_specom,
author={C. S. Kumar and Haizhou Li},
title={{Language identification for multilingual speech recognition systems}},
year=2004,
booktitle={Proc. 9th Conference on Speech and Computer (SPECOM 2004)},
pages={343--347}
}