This paper presents a novel approach of speed control for a permanent magnet synchronous motor (PMSM) using on-line self tuning artificial neural network (ANN). Based on motor dynamics and non-linear load characteristics, an ANN speed controller is developed and integrated with the vector control scheme of the PMSM drive. The combined use of off-line and on-line weights and biases adjustments offers a unique feature of on-line system identification and precise speed control of a high performance PMSM drive. The complete drive system is implemented in real time using a digital signal processor controller board-DS1102 on a laboratory 1 hp PMSM. using the experimental setup, the performances of the proposed drive system are evaluated under various operating conditions. The test results validate the efficacy of the ANN for precise speed control of the PMSM drive. Furthermore, the use of ANN makes the drive system robust, accurate and insensitive to parameter variations.