Method and apparatus for automatically recognizing input audio and/or video streams

Abstract

A method and system for the automatic identification of audio, video, multimedia, and/or data recordings based on immutable characteristics of these works. The invention does not require the insertion of identifying codes or signals into the recording. This allows the system to be used to identify existing recordings that have not been through a coding process at the time that they were generated. Instead, each work to be recognized is “played” into the system where it is subjected to an automatic signal analysis process that locates salient features and computes a statistical representation of these properties. These features are then stored as patterns for later recognition of live input signal streams. A different set of features is derived for each audio or video work to be identified and stored. During real-time monitoring of a signal stream, a similar automatic signal analysis process is carried out, and many features are computed for comparison with the patterns stored in a large feature database. For each particular pattern stored in the database, only the relevant characteristics are compared with the real-time feature set. Preferably, during analysis and generation of reference patterns, data are extracted from all time intervals of a recording. This allows a work to be recognized from a single sample taken from any part of the recording.

What is claimed is: 
     
         1 . Audio signal recognition server apparatus adapted to receive, from a capture device, feature data that corresponds to a captured audio sample that is less than an entire reference audio work, the recognition server apparatus comprising:
 interface structure configured to receive the sample feature data from the capture device;   a memory storing a library comprising (i) a first plurality of reference feature data sets which correspond to a first recorded reference audio work, and (ii) a second plurality of reference feature data sets which correspond to a second recorded reference audio work, each recorded reference audio work being longer than the captured audio sample; and   server processing structure configured to:   receive a first reference input audio signal corresponding to the first recorded reference audio work;   separate the received first reference input audio signal into a first plurality of frequency bands which have different frequencies;   compute the first plurality of reference feature data sets, which correspond to spectrally distinct portions of the first plurality of frequency bands of the first received reference input audio signal;   store in the memory the first plurality of reference feature data sets which correspond to the first reference input audio signal;   receive a second reference input audio signal corresponding to the second recorded reference audio work;   separate the received second reference input audio signal into a second plurality of frequency bands which have different frequencies;   compute the second plurality of reference feature data sets, which correspond to spectrally distinct portions of the second plurality of frequency bands of the second received reference input audio signal;   store in the memory the second plurality of reference feature data sets which correspond to the second reference input audio signal;   compare the sample feature data received by said interface structure with the stored first and second pluralities of reference feature data sets; and   generate a recognition signal in response to the received sample feature data matching at least one reference feature data set of the stored first and second pluralities of reference feature data sets.