Copyright protection for WAV-table synthesis audio using digital watermarking

Abstract

In this paper, we present a novel audio watermarking method for WAV-table synthesis audio (Downloadable Sounds - DLS) copyright protection. In order to guarantee the inaudibility and robustness, we embed watermark both in sample data and in articulation parameters. For sample data, we propose an adaptive low-bit coding method based on finite automaton. For articulation parameters, we generate a virtual instrument and use its parameters to hide watermark information. We also discuss the watermark detection and give the experimental results of watermarked DLS to the listening and robustness tests.