DEVICE AND METHOD FOR RECOGNIZING CONTENT USING AUDIO SIGNALS

摘要

The present invention relates to a device and method for recognizing content using audio signals and includes: a query fingerprint extraction unit that extracts query fingerprints by forming frames that have a preset frame length for an input audio signal and generates a frame-based characteristic vector for each frame; a reference fingerprint database that stores reference fingerprints for comparison with the query fingerprint and content information corresponding to the reference fingerprints; and a fingerprint matching unit that determines the reference fingerprint matched with the query fingerprint, wherein the query fingerprint extraction unit forms a frame while changing the frame moving length and the interval between the start points of adjacent frames in some sections. According to the present invention, it is possible to provide a device and a method for recognizing content using audio signals that quickly provides a result and maintains accuracy and reliability during matching.