[WM]: watermark using the MPEG layer1 psychoacoustic model

["Ronen Cohen, CEM, R&D " <Ronen.Cohen@nice.com>]

Hi all,

I'm trying to use the MPEG layer 1 psychoacoustic model (implemented on MATLAB by Fabien A. P. Petitcolas) in order to embed
watermark in speech signal.

I have the following question:

After shaping the watermark signal using the model I wonder how to embed it into the speech - the model uses overlapping windows of
512 samples with overlap of 128 samples for the FFT calculation. What input samples are to be taken for the embedding process (the
same used for the FFT?) What about the overlapped samples between two consecutive windows are they watermarked twice...?

I probably miss something please tell me what!

 

Any help will be much appreciated!

 

Ronen Cohen


______________________________________________________________________________

Watermarking Mailing List - http://www.watermarkingworld.org/ml.html
To unsubscribe send email to "majordomo@watermarkingworld.org" with
"unsubscribe watermarking YOURMAIL" in the body.
______________________________________________________________________________