This Paper proposed one novel sensorless induction motor speed control system with self-organizing map and fuzzy control. The self-organizing map could train the neural network according to the simulation or experimental induction running parameters automatically. The trained neural network then is employed to preprocess the induction parameters, and the net output is sent to a fuzzy system to adaptively adjust the speed regulation. This method could increase the steady and dynamic response of the sensorless induction motor control system, which has been demonstrated by the simulation results.