SOUND PROCESSING METHOD, SOUND PROCESSING SYSTEM, VIDEO PROCESSING METHOD, VIDEO PROCESSING SYSTEM, SOUND PROCESSING DEVICE, AND METHOD AND PROGRAM FOR CONTROLLING SAME

摘要

Provided is a device capable of real-time sound identification and matching by solving both the problem of reducing the time length of a frame and improving temporal accuracy and the problem of being robust against mixing with other sounds. This sound processing device, which generates a sound identifier for identifying a sound on the basis of time-frequency analysis of a sound signal, comprises: a time-frequency analyzer that generates a time-frequency plane from a sound signal through time-frequency analysis; a region characteristic-amount extractor that, for a plurality of partial-region pairs defined on the generated time-frequency plane and each consisting of two partial regions that differ from one another in shape and/or position, extracts a region characteristic amount from each partial region; and a sound identifier generator that, by using the region characteristic amount extracted from each partial region, generates a sound identifier for identifying the sound.