Cognitive model for speech perception mechanism on the base of multiagent systems

This paper proposes the formal description of the structure of an act of speech perception, which is necessary as a general theoretical basis for the development of universal automatic speech recognition systems of high performance in real operation conditions and cocktail party situations. The general structural dynamics of the speech recognition process has been developed. The necessity of using the articulation event as a minimal basic pattern of sound image recognition has been proved. Multi-agent systems were chosen as the formal means of implementation.