1st International ICST Workshop on IMS and Mobile Multimedia

Research Article

A SPIT Detection Mechanism Based on Audio Analysis

Download502 downloads
  • @INPROCEEDINGS{10.4108/ICST.MOBIMEDIA2008.3850,
        author={Yacine Rebahi and Sven Ehlert and Andreas Bergmann},
        title={A SPIT Detection Mechanism Based on Audio Analysis},
        proceedings={1st International ICST Workshop on IMS and Mobile Multimedia},
        publisher={ICST},
        proceedings_a={IMS},
        year={2010},
        month={5},
        keywords={Spam SPIT audio analysis speech processing signal processing},
        doi={10.4108/ICST.MOBIMEDIA2008.3850}
    }
    
  • Yacine Rebahi
    Sven Ehlert
    Andreas Bergmann
    Year: 2010
    A SPIT Detection Mechanism Based on Audio Analysis
    IMS
    ICST
    DOI: 10.4108/ICST.MOBIMEDIA2008.3850
Yacine Rebahi1,*, Sven Ehlert2,*, Andreas Bergmann3,*
  • 1: Fraunhofer Fokus, Kaiserin Augusta Allee 31, 10589 Berlin, Germany. +49 30 34637141
  • 2: Fraunhofer Fokus, Kaiserin Augusta Allee 31, 10589 Berlin, Germany. +49 30 34637378
  • 3: Eleven GmbH, Hardenbergplatz 2, 10623 Berlin, Germany. +49 30 520056159
*Contact email: yacine.rebahi@fokus.fraunhofer.de, sven.ehlert@fokus.fraunhofer.de, Andreas.Bergmann@eleven.de

Abstract

The Session Initiation Protocol (SIP) provides new means for establishing and maintaining IP based multimedia services. Similar to most of the Internet protocols, SIP might suffer from the spam problem. The latter refers to sending unsolicited information in bulk. In this paper, we describe a way of how analysis of the audio content can help to detect spam calls. After identifying the requirements for the audio analysis to be successful in detecting spam, an architecture suited for VoIP media is suggested. To prove the feasibility of our solution, the audio analysis technique is validated through various testing methods.