|
Audiomatic Identification (AudioID) via acoustic fingerprints
The basic concept behind a fingerprinting system is to identify a piece of audio content by extracting a compact and unique signature from it (the so called fingerprint). Depending on the required reliability such a fingerprint has a size from appr. one to four kB/minute. This scalability enables a flexible compromise between the compactness (filesize) and robustness.
Fingerprints must be computed from a set of known audio material and stored in a database referencing the original metadata. Unknown content can then be identified by comparing its fingerprint to the ones contained in the database. Thereby, it is possible to identify the audio data even if there is just a short piece of it available. The basic technology is part of the international ISO/IEC MPEG-7 
audio standard of the Moving Picture Experts Group (MPEG).
The AudioID technology has been jointly developed
at the Fraunhofer Institute for Integrated Circuits IIS and
the Fraunhofer Institute for Digital Media Technology IDMT.
For this achievement and for the first successful market launch,
the researchers were honoured with the Joseph-von-Fraunhofer
Award in 2004.
However, both institutes are no longer responsible
for licensing issues. If you want to learn more about licensing
conditions and prices please visit www.mufin.com or contact:
mufin GmbH
Friedrichstraße 200
10117 Berlin
Germany
Tel.: +49 (0) 30 / 29392 – 454
Fax: +49 (0)30 / 29392 – 400
info@mufin.com |
