Dear,

        Ahmed Hussain Ali

https://archive.org/details/defreville-Aucouturier_urbanDb

this link will give you a audio data set according to your given format 

it might be helpful for you

Hi,

Not directly related to your question, but to your general problem domain:

A couple of years ago, I took a Specialization degree on Information Security, and as a research project, I presented a "Model for covert channel evaluation for establishing secure communications". My text is in Spanish, but if you think it could be interesting to you, I'd be more than happy in explaining its core ideas to you.

You can download my thesis at the link I present next.

http://ru.iiec.unam.mx/2677/

Dear Ahmad

Could you be more spesific, what kind data set do you need ? because if you talk about steganography in WAV audio files you could use and random data and implement at signal noise