Audio watermarking has been proved as a possible solution to avoid illegal usage of audio files. It embeds copyright information into audio files as a proof of their ownership. This paper deals with the design a system which is capable of hiding digital data inside an audio signal. Digital data which is to be embedded can be binary sequence, text or image. The proposed method can embed the watermark data in approximation coefficients of discrete wavelet transform. Experimental results for different audio signals show that this watermarking technique is robust against the common signal processing attacks such as resampling, Requantization, Low pass filtering, Volume scaling & Noise addition.