A two-level audio fingerprint retrieval algorithm for advertisement audio

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

Abstract

In this paper, we propose a two-level audio fingerprint retrieval algorithm to satisfy the demand of accurate and efficient search for advertisement audio. We use the spectrum modulus to extract two kinds of audio fingerprints from global and local view of the audio frame, and then we carry on a two-level advertisement audio search. Finally, we parallel the search by MPI to enhance the search speed and make our system have a good scalability.

Original languageEnglish
Title of host publication12th International Conference on Advances in Mobile Computing and Multimedia, MoMM 2014
EditorsYu-Hui Tao, Hsin-Chang Yang, I-Hsien Ting, Matthias Steinbauer, Ismail Khalil, Gabriele Anderst-Kotsis
PublisherAssociation for Computing Machinery
Pages235-239
Number of pages5
ISBN (Electronic)9781450330084
DOIs
StatePublished - 8 Dec 2014
Event12th International Conference on Advances in Mobile Computing and Multimedia, MoMM 2014 - Kaohsiung, Taiwan, Province of China
Duration: 8 Dec 201410 Dec 2014

Publication series

Name12th International Conference on Advances in Mobile Computing and Multimedia, MoMM 2014

Conference

Conference12th International Conference on Advances in Mobile Computing and Multimedia, MoMM 2014
Country/TerritoryTaiwan, Province of China
CityKaohsiung
Period8/12/1410/12/14

Keywords

  • Advertisement audio retrieval
  • Audio fingerprints
  • MPI
  • Spectrum modulus
  • Two-level

Fingerprint

Dive into the research topics of 'A two-level audio fingerprint retrieval algorithm for advertisement audio'. Together they form a unique fingerprint.

Cite this