Abstract
Audio dual compression detection is an important part of audio forensics. It is of great significance to judge whether the audio has been falsified and forged. This study found that the advanced audio coding (AAC) audio scale factor gradually decreases with the number of compressions increases. Based on this, we propose an AAC double compression audio detection algorithm based on the statistical characteristics of the scale factor difference before and after audio re-compression. The experimental results show that the algorithm can accurately classify dual compressed AAC audio. The average accuracy of AAC audio classification between low-bit-rate transcoding to high-bit-rate is 99.91%, and the accuracy rate between the same bit rate is 97.98%. In addition, experiments with different durations, different noises, and different encoders also proved the better performance of this algorithm.
Funding Information
  • National Natural Science Foundation of China (Grant No. U1736215, 61672302)
  • Zhejiang Natural Science Foundation (Grant No. LZ15F020002, LY17F020010)
  • Natural Science Foundation of Ningbo Municipality (Grant No. 2017A610123)
  • Ningbo University Fund (Grant No. XKXL1509, XKXL1503)

This publication has 16 references indexed in Scilit: