An Adaptive Audio Watermarking Algorithm Based on Cepstrum Transform

Abstract

It is needed to improve the poor invisibility and maintain its robustness of the watermarking embedded in the silent or faint sound clip of the audio files. The audio signal energy size determines whether the watermarking is embedded in the location. After dividing the audio files into several parts, the suitable parts are selected to embed watermarking based on cepstrum transform ,according to the HAS(human auditory system). Proper selection of cepstrum coefficient the watermarking be of strong anti-attack ability and anti-jamming ability. Using auditory masking properties make the watermarking be of consistent good invisibility. The algorithm combined psychoacoustic model with cepstrum transform can be adaptive to embed watermarking in the complicated audio files.