The invention belongs to the technical field of wind power generator fault diagnosis, and particularly relates to a wind power generator set mechanical fault audio recognition and fault diagnosis method which comprises fault signal detection and a convolutional neural network. The fault signal detection comprises vibration signal detection, acoustic emission signal detection, strain force signal detection, temperature signal detection, oil parameter detection and electric signal detection. A set of method flow based on vibration fault signal monitoring and a convolutional neural network model is provided for intelligent bearing fault diagnosis, a vibration signal is obtained by an acceleration sensor, historical data is subjected to reasonable sampling and 1D-2D signal processing conversion, and then the vibration signal is obtained by the acceleration sensor. Bearing signal samples in various fault states are reasonably divided into a training set and a test set, the training set is sent to the established deep convolutional neural network for model learning, and after model learning is completed, the test set is used for verifying the generalization ability of the model, namely the test accuracy.