Establishing the authenticity of an audio evidence is an important task in a criminal investigation. Apart from the actual signal content, the audio evidence may have artifacts due to background noise or reverberation. Inconsistencies in these artifacts can provide clues about any malicious attacks on the recording. In this paper, a methodology to detect tampering in the audio recording using the reverberation component has been proposed. A threshold based clustering technique is used to identify any unauthorized alteration on the evidentiary file. The clusters depict whether the group of audio samples belong to the same acoustic environment or are from a different recording recorded in a different environment. On the basis of number of clusters obtained, tampering can be detected.
(See document for exact affiliation information.)
Click to purchase paper as a non-member or login as an AES member. If your company or school subscribes to the E-Library then switch to the institutional version. If you are not an AES member Join the AES. If you need to check your member status, login to the Member Portal.
