In the paper the model of voice interface, which provides automatic input of Russian speech is proposed. The morphemes level of speech representation is introduced and as a result the size of vocabulary is significantly decreased. The developed morphemes databases are used for collecting the statistics of morphemes co-ordination by text corpuses. At that during recognition the degree of co-ordination between root morphemes has main significance. As a result of such processing the invariance to grammatical deviations is provided and also the speed of recognition of Russian speech and other languages with complex mechanism of word formation is improved.
Cite as: Ronzhin, A.L., Karpov, A.A. (2004) Implementation of morphemic analysis for Russian speech recognition. Proc. 9th Conference on Speech and Computer (SPECOM 2004), 291-296
@inproceedings{ronzhin04_specom,
author={Andrey L. Ronzhin and Alexey A. Karpov},
title={{Implementation of morphemic analysis for Russian speech recognition}},
year=2004,
booktitle={Proc. 9th Conference on Speech and Computer (SPECOM 2004)},
pages={291--296}
}