An Approach to Digital Watermarking of Speech Signals in the Time-Frequency Domain

A watermarking scheme in the time-frequency domain is proposed. The region for watermark embedding is selected by using the S-method based on the time-frequency representation. Time-varying filter scheme is used to map the watermark sequence from the time-frequency domain to the time domain. Watermark detection is performed in the time-frequency domain. Theory is illustrated by the example.