Detection of tampering in lossy compressed digital audio recordings

This paper addresses the problem of tampering detection and discusses one of the methods used for authenticity analysis of digital audio recordings. Presented approach is based on checking frame offsets in audio files compressed by using perceptual audio coding. This method can be further improved by applying histogram analysis of modified discrete cosine transform spectrum and detection of maximum number of nonzero spectral coefficients. The influence of compression algorithms' parameters on detection of forgeries are presented by applying AAC and Ogg Vorbis encoders as examples. The effectiveness of tampering detection algorithms proposed in this paper is tested on a predefined music database and compared graphically using ROC-like curves.