Raw Speech Watermarking using Wavelet Packet Transform

P. Tungtivachagul and T. Amornraksa (Thailand)


Digital signal processing, Speech recognition, Digitalwatermarking, Wavelet packet transform.


This paper presents a concept of using raw speech signal as a watermark signal to enhance the robustness of the embedded watermark within the watermarked data such as a video sequence. This idea arises from the fact that the raw speech signal contains a considerable amount of redundancies, and its contents can still be recognized after retrieval process, even if the watermarked data was badly attacked. As long as the raw speech signal, extracted from the attacked watermarked video, contains enough important information, its contents can still be intelligible. Moreover, several video compression schemes presently apply wavelet-based techniques for reducing the redundancies contained within the data. The raw speech signal is therefore embedded into the video sequence by a wavelet-based technique i.e. wavelet packet transform. The results from experiments show the impressive quality of the watermarked image. Furthermore, when the embedded speech is extracted, after being attacked by some common signal processing, the listening test is performed to evaluate the intelligibility contained in the extracted speech. The impression of the human perceptual system, as it tends to adjust and learn quickly to recognize the repeated speech, enhances the probability of recognizing the contents of the extracted raw speech signal precisely.

Important Links:

Go Back