Audio Amplitude-Level Quantification Vector for Identification of Audio Post-Processing Operation

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.