Robust AVS audio watermarking

Abstract

Part III of AVS (China Audio and Video Coding Standard) is the first generic audio standard proposed in China and is becoming popular in some applications. For MP3 audio, some efforts have been made to solve the copyright pirating and malicious modifications by the way of watermarking. To our best knowledge, similar work on the AVS audio has not been explored. In this paper, we present a novel robust watermarking algorithm for AVS audio. We find that the frame type in the AVS coded process can be used as a robust feature for watermarking. Then we design a novel watermarking algorithm which is integrated into the AVS audio coding process. The watermark bits are embedded into the compressed bit stream. At the extracting end, the watermark bits can be extracted from the compressed bit stream directly. This algorithm achieves robustness to decoding/re-coding attacks, and has low computation complexity for both embedding and extracting while preserving the quality of the audio signals.

AC-3 bit stream watermarking

“…Instead of separate compression and watermarking, combined approaches have been proposed in [6] and [7]. These techniques take Puls Code Modulation (PCM) samples as input for the embedder and generate watermarked mp3 or Audio Video Coding Standard (AVS) bit streams as output.…”