Mixed Watermarking-Fingerprinting Approach for Integrity Verification of Audio Recordings

Abstract— W e introduce a method for audio-integrityverification basedon a combination of watermarking and finger printing. An audio finger -print is a perceptual digest that holds content information of a recordingand allows one to identify it from other recordings. Integrity verificationis performed by embedding the finger print into the audio signal itself bymeans of a watermark. The original finger print is reconstructed from thewatermark and compared with a new finger print extracted from the ob-served signal. If they are identical, the signal has not been modified; if not,the system is able to determine the approximate locations where the signalhas been corrupted. The watermarked signal may undergo content pre-serving transformations, such as resampling or D/A and A/D conversion,without triggering the corruption alarm. I. I NTRODUCTION using N many applications, the integrity of an audio recording mustbe unquestionably established before the signal can actuallybe used, i.e. one must be sure that the recording has not beenmodified without authorization.When dealing with speech, some application contexts inwhich integrity must be ensured are: protection of previously recorded testimonies that are to beused as evidence before a court of law; protection of recorded interviews, which could be edited formalicious purposes.Regarding music applications, some examples are: integrity verification of radio or television commercials to en-sure they air as negotiated;

[1]  Pedro Cano,et al.  Automatic Segmentation for Music Classification using Competitive Hidden Markov Models , 2000, ISMIR.

[2]  Graham Shaw,et al.  Digital document integrity , 2000, MULTIMEDIA '00.

[3]  Ronald L. Rivest,et al.  The MD5 Message-Digest Algorithm , 1992, RFC.

[4]  Ahmed H. Tewfik,et al.  Digital watermarks for audio signals , 1996, 1996 8th European Signal Processing Conference (EUSIPCO 1996).

[5]  Markus G. Kuhn,et al.  Information hiding-A survey : Identification and protection of multimedia information , 1999 .

[6]  Ralf Steinmetz,et al.  Content-based digital signature for motion pictures authentication and content-fragile watermarking , 1999, Proceedings IEEE International Conference on Multimedia Computing and Systems.

[7]  C.-C. Jay Kuo,et al.  Speech content integrity verification integrated with ITU G.723.1 speech coding , 2001, Proceedings International Conference on Information Technology: Coding and Computing.

[8]  Jana Dittmann Content-fragile watermarking for image authentication , 2001, IS&T/SPIE Electronic Imaging.

[9]  Rosario Gennaro,et al.  How to Sign Digital Streams , 1997, CRYPTO.

[10]  Markus G. Kuhn,et al.  Information hiding-a survey , 1999, Proc. IEEE.

[11]  Marcos Perreau Guimaraes,et al.  Optimisation de l'allocation des ressources binaires et modelisation psychoacoustique pour le codage audio , 1998 .