Speech recognition based on word identifiers possibilities research

• Autorzy: Śliwiński, A.
• Języki publikacji: EN

Abstrakty

EN

The article describes the results of preliminary research concerning speech commands recognition, carried out in order to test their potential use in simple, low computing power, built-in systems. In the course of the research, both single-pattern command and double-pattern (male and female) command results have been compared. Perfunctorily-devised word identifiers have been used in order to speed up calculations. The identifiers are based on results from the Fast Fourier Transform (FFT). Double-pattern use has been checked in order to improve command identification quality.

• Czasopismo: Zeszyty Naukowe. Elektryka / Politechnika Opolska
• Rocznik: 2016
• Tom: z. 72
• Strony: 91--92
• Opis fizyczny: Bibliogr. 2 poz., tab.

Twórcy

autor

Śliwiński, A.

• Faculty of Electrical Engineering, Automatic Control and Informatics Institute of Electromechanical Systems and Industrial Electronics

Bibliografia

• [1] ŚLIWIŃSKI A., TOMCZEWSKI K.: Autonomous control systems speech recognition possibilities research, Poznan University of Technology Academic Journals Electrical Engineering, 2016
• [2] WALENDOWSKI P.: Application of SVM neural networks for speech recognition Dissertation, Politechnika Wrocławska, 2008