Generating a reference audio fingerprint for an audio signal associated with an event

摘要

An audio identification system generates a reference audio fingerprint associated with an event. The reference audio fingerprint is generated from samples of an audio signal associated with the event captured by multiple devices. To generate the reference audio fingerprint, fingerprints are generated from each sample, and the generated fingerprints are temporally aligned. Fingerprints associated a temporally overlapping portion of the audio signal are averaged, and the average value is associated with the temporally overlapping portion of the audio signal and included in the reference audio fingerprint. The reference audio fingerprint is stored along with identifying information, such as an event name, an event time, an event date, or other information describing the event associated with the audio signal from which the samples were captured.

主权项

1. A computer-implemented method comprising:
obtaining a plurality of samples of one or more audio signals from a live event attended by a plurality of users, each sample representing a portion of an audio signal from the live event and being associated with identifying information for the live event; identifying a plurality of samples of an audio signal from the live event based on the identifying information associated with the samples; generating a set of audio fingerprints based on the identified plurality of samples; combining audio fingerprints of the set of audio fingerprints into a temporally aligned series; generating a reference audio fingerprint associated with the live event by averaging audio fingerprints in the temporally aligned series associated with overlapping time intervals, where the reference audio fingerprint has a reduced noise compared to audio fingerprints of the set of audio fingerprints; and storing the reference audio fingerprint and an association between the reference audio fingerprint and information describing the live event.