A machine learning model for audio detection of Capuchin bird. Here in this project I build a project of classifying the audio of a Capuchin bird. First we train the model on our Capuchin bird clips that are 2 to 3 seconds long. Here we have a deep learning model with CNN and ANN architecture in use. We divided our forest clips which were around 3 minutes long into several fragments to calculate the density of capuchin bird clips. We test our model on each fragment and try to find whether that fragment has the capuchin bird clips or not. After making the results of each forest clips we make a CSV file of the results. You can download the dataset from https://www.kaggle.com/datasets/kenjee/z-by-hp-unlocked-challenge-3-signal-processing