Skip to Main Content
ASME Press Select Proceedings

International Conference on Computer and Electrical Engineering 4th (ICCEE 2011)

By
Jianhong Zhou
Jianhong Zhou
Search for other works by this author on:
ISBN:
9780791859841
No. of Pages:
698
Publisher:
ASME Press
Publication date:
2011

The main aim of this paper is to carry out automatic identification of Indian languages using vector quantization. In this particular work we have applied vector quantization classification technique on LPC derived features for identifying 3 Indian languages Assamese, Bengali and Indian English. Experimental results show that recognition accuracy for Assamese & Indian English is better compared to Bengali language. Stop words were used for reducing the overhead of testing process and it shows that Assamese gives optimal recognition rate at 67 % and Bengali, Indian English gives 100% recognition rate. It has also been observed here from mean square error that quality of speech signal used here is very good.

This content is only available via PDF.
You do not currently have access to this chapter.
Close Modal

or Create an Account

Close Modal
Close Modal