Abstarct: The name " speech enhancement '' refers to large group of methodes that improve the quality and intelligibility of noisy speech by suppressing the background noise from noisy signal. There are a lot of methodes and papers about speech enhancement. In this paper,we propose a new speech enhancement system using the wavele neural network. Wavenet is a special feedforward neural network, which uses wavelet basis function as an activation function. The wavelet coefficients for the specific signal processing problems are obtained by updating the weights of the adaptive Wavenet using the coinjugate gradient methode.WNN uses a feedforward network with a learning algorithm that optimizes the network parameters in such a way that the mean squared error (MSE) between the desirable signal and the output signal is minimized. Daubechies (db5) mother function is used in the estimation of noise .The proposed methode was evaluated on several speakers and under various noise conditions including white Gaussian noise , babble noise and F16. Finally, the proposed algorithm is evaluated in term of SNR, Segmental SNR and LLR(Log Likellihood Ratio) and then the proposed methode is compared with neural network and composite of neural network and wavelet and wavelet methodes. The results of this comparison and experimental results show that the proposed methode has an acceptable performance.