Acoustic fingerprint

From Wikipedia, the free encyclopedia
Jump to: navigation, search
For acoustic emissions of ships and submarines, see Acoustic signature.

An acoustic fingerprint is a condensed digital summary, deterministically generated from an audio signal, that can be used to identify an audio sample or quickly locate similar items in an audio database.[1]

Practical uses of acoustic fingerprinting include identifying songs, melodies, tunes, or advertisements; sound effect library management; and video file identification. Media identification using acoustic fingerprints can be used to monitor the use of specific musical works and performances on radio broadcast, records, CDs and peer-to-peer networks. This identification has been used in copyright compliance, licensing, and other monetization schemes.

Attributes[edit]

A robust acoustic fingerprint algorithm must take into account the perceptual characteristics of the audio. If two files sound alike to the human ear, their acoustic fingerprints should match, even if their binary representations are quite different. Acoustic fingerprints are not bitwise fingerprints, which must be sensitive to any small changes in the data. Acoustic fingerprints are more analogous to human fingerprints where small variations that are insignificant to the features the fingerprint uses are tolerated. One can imagine the case of a smeared human fingerprint impression which can accurately be matched to another fingerprint sample in a reference database; acoustic fingerprints work in a similar way.

Perceptual characteristics often exploited by audio fingerprints include average zero crossing rate, estimated tempo, average spectrum, spectral flatness, prominent tones across a set of bands, and bandwidth.

Most audio compression techniques (AAC, MP3, WMA, Vorbis) will make radical changes to the binary encoding of an audio file, without radically affecting the way it is perceived by the human ear. A robust acoustic fingerprint will allow a recording to be identified after it has gone through such compression, even if the audio quality has been reduced significantly. For use in radio broadcast monitoring, acoustic fingerprints should also be insensitive to analog transmission artifacts.

On the other hand, a good acoustic fingerprint algorithm must be able to identify a particular master recording among all the productions of an artist or group. For use as evidence in a court of law, an acoustic fingerprint method must be forensic in its accuracy.[citation needed]

Implementations[edit]

This is a list of notable acoustic fingerprinting products.

Proprietary
  • All Media Guide's LASSO is a commercial service that uses acoustic fingerprinting, and other techniques, to recognize music. (U.S. Patent 7,277,766)
  • Audible Magic Corporation is a commercial venture that provides electronic media identification and copyright management solutions using proprietary acoustic fingerprinting technology U.S. Patent 5,918,223 based on original research by Muscle Fish Consulting[2]
  • AudioID is a commercial technology for automatically identifying audio material using acoustic fingerprints. It was developed by the German Fraunhofer Institute.
  • Axwave's Gsound uses acoustic fingerprinting to identify music, film and TV from cell phone or from computers. It work also by sending to the servers simple wave files.
  • BAS - Broadcasting Airplay System is the algorithm of digital fingerprint, for which Knowmark is owner. It is used in Italy for dance halls songs recognition and for radio broadcasters, both music and commercials. It is also used in the innovative platform Mira-PMM for measurement of cross-media audience. Italy patent no. 0001393620
  • BMAT Vericast is a global music identification service that monitors millions of songs over 2000 radios and televisions across more than 50 countries worldwide. The solution provides real time recognition and auditable reporting based on an audio fingerprint that is resistant to signal alterations such as voice over, broadcast mastering or noisy channel degradation.
  • YouTube's Content ID is able to identify audio/visual part of copyrighted content. In case if whole third party content is matched or only certain portion, user can check whether the content is properly identified or misidentified.
  • AudioPrint is a commercial technology, based on the variations of spectral sub-band energies. It was developed by the French Research Institute IRCAM, during the Quaero project (see [1]).
  • Gracenote's MusicID is a commercial product that uses acoustic fingerprinting along with other methods to identify music.
    • The Nero Multimedia Suite Nero (software suite) version 9 and 10 uses Gracenote to add metadata like author, title and genre to an audio file.
    • Sony Ericsson's TrackID software uses Gracenote to identify songs being recorded via cell phone in a way similar to Shazam.
    • Winamp version 5.5 uses Gracenote to power automatic playlist generation with "Nullsoft Playlist Generator" plugin that comes with the software.
  • Midomi is a commercial service that can match music clips, as well as identifying a song that is sung or hummed
  • Moodagent is a commercial service from Syntonetic that combines digital signal processing and AI techniques to create music profiles that incorporate characteristics such as mood, emotion, genre, style, instrument, vocals, orchestration, production, and beat/tempo.
  • SoundHound, an acoustic fingerprint-based service with web and mobile platforms (Android, iOS, Windows Phone) that allows songs or hummed tunes to be identified using the Midomi service.
  • Shazam, an acoustic fingerprint-based service allows for songs to be identified via cell phone.
  • Tunatic by Wildbits is an application that allows identifying music while being played, analyzing the songs and comparing with the information on a server
Open source
  • MusicBrainz, a free and open content project for a music database that uses AcoustID's free database of audio fingerprints, which aims to map its fingerprints to the MusicBrainz database. MusicBrainz also used MusicIP's Open Fingerprint Architecture for fingerprinting and the AmpliFIND (formerly MusicDNS) service for identifying audio files since 2006,[3] but is phasing out AmpliFIND in favour of the open source AcoustID, after AmpliFIND was acquired by Gracenote in 2011.[4]
  • Last.fm's own acoustic fingerprinting application[5] was released in 2007.[6] The technology is now included in the Last.fm client software.
  • AcoustID is an open source project that aims to create a free database of audio fingerprints with mapping to the MusicBrainz metadata database and provide a web service for audio file identification using this database.
  • Echoprint is an open source music fingerprint and resolving framework powered by the The Echo Nest.

See also[edit]

References[edit]

  1. ^ ISO IEC TR 21000-11 (2004), Multimedia framework (MPEG-21) -- Part 11: Evaluation Tools for Persistent Association Technologies
  2. ^ "Content-Based Classification, Search, and Retrieval of Audio," IEEE MultiMedia, vol. 3, no. 3, pp. 27-36, Sept., 1996.
  3. ^ Robert Kaye (mayhem) (12 March 2006). "New fingerprinting technology available now!". MusicBrainz Blog. 
  4. ^ Picard 0.16 released with AcoustID support
  5. ^ lastfm/Fingerprinter · GitHub
  6. ^ Last.fm blog

US patent 7346472, Moskowitz, et al., "Method and device for monitoring and analyzing signals ", issued March 18, 2008 "signal abstract"

External links[edit]