Application of Double Pictures to Audio Watermarking using Discrete Cosine Transform (DCT) and Singular Value Decomposition (SVD) Methods

Abstract

One way to control the development of arbitrary distribution of audio is to apply audio watermarking. Audio Watermarking is used to identify the ownership rights of audio files by embedding a watermark while paying attention to the audio quality and the watermark's resistance to amplify, resampling, and invert attacks. Discrete Cosine Transform (DCT) and Singular Value Decomposition (SVD) methods are frequency-domain methods that have a good robustness. The DCT and SVD methods are combined with inserting and extracting watermarks using two images as watermarks. From the test results, the effect of inserting an audio watermark with two images does not significantly impact the durability of the watermark and the quality of the audio inserted with the watermark. However, the alpha value has a significant effect. The higher the alpha value, the lower the audio SNR value by an average of 1.93 dB from an alpha value of 0.0008 and 0.001 for each type of classical, jazz, and rock audio genre.