AUDIO SIGNAL ANALYSIS METHOD, VOICE RECOGNITION METHODS USING SAME, AND THEIR DEVICES, PROGRAM, AND RECORDING MEDIUM THEREOF

摘要

PROBLEM TO BE SOLVED: To suppress influences of addition distortions by solving the problem, in extraction of parameters that express the characteristics of signals, including not only voice signals but also periodic components, such as music signals. SOLUTION: An audio signal analysis method comprises dividing a voice signal into a plurality of band signals by a filtering bank 11; finding non-periodic component powers (13, 14 and 15A) and periodic component powers (15F and 16) of each band signal; averaging each noise section of the periodic component power and the non-periodic component power with time average means 21P and 21A over a long time, respectively; subtracting the periodic components and the non-periodic component powers in the long-time average of the periodic components and the non-periodic component, respectively; converting subtraction results of the periodic components and non-periodic components into discrete cosines, respectively (18P and 18A); and vector-coupling at least a part in a manner of each hour and/the order of the periodic components and non-periodic component discrete cosine coefficients. COPYRIGHT: (C)2006,JPO&NCIPI