REVIEW AND CLASSIFICATION METHODS FOR PROCESSING SPEECH SIGNALS IN THE SPEECH RECOGNITION SYSTEMS

Alimuradov Alan Kazanferovich, applicant, sub-department of information and measuring equipment and metrology, Penza State University, alansapfir@yandex.ru
Churakov Petr Pavlovich, doctor of technical sciences, professor, sub-department of information and measuring equipment and metrology, Penza State University, iit@pnzgu.ru

004.934:681.391

Background. Work in the field of speech signals processing is quite active. The article presents an analytical review and classification of the existing processing methods applied in speech recognition systems. Materials and methods. A review is made on the basis of an analytical study of processing methods that use different types of speech signals analysis. Results. The classification of processing techniques in frequency, temporary and temporary-frequency areas used in filtering, segmentation, detection and recognition of speech signals is presented. Conclusions. The review and the presented classification allow to objectively estimate the capabilities of existing processing methods, to evaluate the potential application of new mathematical apparatus in speech signals processing tasks in speech recognition systems.

speech signal processing, speech recognition, voice control.