Speech Recognition using Convolution neural network, is used to recognized the words and digitize them and analyze the sound. It trains a deep learning model that detects the presence of speech commands in audio it implemented by using MATLAB. It uses a convolution neural network to train a model. The model was trained for commands and background noise. The trained model getting accuracy of 96.34% while testing the data. Define the level for audio processing and the level of identification in Hz and build an audio interface viewer that can interpret audio from your microphone. When we speak commands it detects and visualize it and we speak other than commands it shows unknown. When we not speaking anything it detecting background noise.