Digital audio watermarking method using hybrid transform

The present invention relates to a digital audio watermarking method using a hybrid conversion, the audio signal is DCT (discrete cosine transform) and then set the threshold value by analyzing the energy magnitude of the coefficients, by taking the coefficients above the threshold tonal signals (tonal Remove the signal), and constitutes a nasal action signal (nontonal signal) in the remaining coefficients. Separated into tonal signal and the audio signal is non-negative signal crude inserts the watermark by a hybrid transform using the DCT and DWT (discrete wavelet transform), depending on their properties. That is, the number of low-frequency components into tonal signal composed inserts a watermark through the QIM (quantization index modulation) method in the DCT domain. The variation between samples many nasal tank signal may be represented by a number of wavelet coefficients value having a value larger inserts a watermark through the QIM method in the DWT domain. Watermarking (watermarking), DCT (discrete cosine transform), DWT (discrete wavelet transform), a hybrid conversion (hybrid transform), QIM (quantization index modulation)