TY - GEN
T1 - Audio amplitude-level quantification vector for identification of audio postprocessing operation
AU - Chen, Zekun
AU - Yuan, Xiaochen
N1 - Publisher Copyright:
© 2018 IEEE.
PY - 2018/7/2
Y1 - 2018/7/2
N2 - Audio tampering is typically followed by post-processing operations to mask the artifacts potentially perceptible by human ears and blur the traces of tampering. However, research on the issue of audio post-processing identification is still a blanket. This paper mainly introduces a method to identify audio post-processing operations. A new audio feature - Audio Amplitude-Level Quantification Vector (AQV) is proposed, then the probability distributions of AQV of audio are calculated and extracted as audio features which are then used for identification of various audio processing. During the detection, the K-Nearest Neighbors (KNN) classifier is applied for classification. Experimental results show that the proposed AQV method can not only verify the authenticity of the speech audio, but also have a significant effect on identifying different types of post-processing operations.
AB - Audio tampering is typically followed by post-processing operations to mask the artifacts potentially perceptible by human ears and blur the traces of tampering. However, research on the issue of audio post-processing identification is still a blanket. This paper mainly introduces a method to identify audio post-processing operations. A new audio feature - Audio Amplitude-Level Quantification Vector (AQV) is proposed, then the probability distributions of AQV of audio are calculated and extracted as audio features which are then used for identification of various audio processing. During the detection, the K-Nearest Neighbors (KNN) classifier is applied for classification. Experimental results show that the proposed AQV method can not only verify the authenticity of the speech audio, but also have a significant effect on identifying different types of post-processing operations.
KW - Audio Amplitude-Level Quantification Vector (AQV)
KW - Audio Feature
KW - Audio Post-processing Detection
KW - K-Nearest Neighbors (KNN)
UR - http://www.scopus.com/inward/record.url?scp=85062230950&partnerID=8YFLogxK
U2 - 10.1109/SNSP.2018.00050
DO - 10.1109/SNSP.2018.00050
M3 - Conference contribution
AN - SCOPUS:85062230950
T3 - Proceedings - 2018 International Conference on Sensor Networks and Signal Processing, SNSP 2018
SP - 226
EP - 230
BT - Proceedings - 2018 International Conference on Sensor Networks and Signal Processing, SNSP 2018
PB - Institute of Electrical and Electronics Engineers Inc.
T2 - 1st International Conference on Sensor Networks and Signal Processing, SNSP 2018
Y2 - 28 October 2018 through 31 October 2018
ER -