RUS  ENG
Full version
JOURNALS // News of the Kabardino-Balkarian Scientific Center of the Russian Academy of Sciences // Archive

News of the Kabardin-Balkar scientific center of RAS, 2017 Issue 4, Pages 11–18 (Mi izkab159)

COMPUTER SCIENCE. MATHEMATICS

Automatic continuous speech recognition system based on multiagent recursive cognitive architectures

Z. V. Nagoev, I. A. Gurtueva

Institute of Computer Science and Problems of Regional Management – branch of Federal public budgetary scientific establishment "Federal scientific center "Kabardin-Balkar Scientific Center of the Russian Academy of Sciences", 360000, KBR, Nalchik, 37-a, I. Armand St.

Abstract: The article presents a concise overview on current approaches to problems of speech recognition. Authors demonstrate the necessity to develop fundamentally new mathematical methods. The concept for fundamentally new approach to developing speech recognition systems is proposed. Speech Recognition Architecture in a medium with several speakers is built on the base of multiagent recursive cognitive models. It imitates human’s attention mechanism. Experimental results are represented.

Keywords: speech recognition, hidden Markov modeling, artificial intelligence, speech processing, multiagent systems.

Received: 21.08.2017



© Steklov Math. Inst. of RAS, 2024