This paper comprehensively investigates the impact of zero-amplitude sample padding during the transcoding of lossy compressed Advanced Audio Coding (AAC) streams. The study was conducted under controlled conditions, using test data from an iPhone 12 Pro Max (v17.0.3) to ensure the reliability of the findings. It explores zero-amplitude sample padding across varying background noise levels and transcoded audio using multiple software applications and versions. The study quantifies the influence of zero-amplitude sample padding in derivative audio files, revealing its presence across diverse noise conditions, with notable variations observed at the beginning of the transcoded audio signal. The analysis of transcoding software underscores the variability in padding length and its effect on audio stream hashing. This research significantly contributes to understanding the impact of zero-amplitude sample padding for audio processing algorithms and forensic analysis. It drives future research on temporal changes in audio signals and digital signal processing. Recommendations for further research include exploring temporal changes in audio signals and optimizing digital signal processing algorithms to accommodate zero-amplitude sample padding variability.
(See document for exact affiliation information.)
https://aes2.org/publications/elibrary-page/?id=22626
(555KB)
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.
