This paper aims to present a neural network based on sequence learning for speech recognition. Since, the sequence learning is an important component of learning in many task domains of intelligent system such as, speech recognition, natural language processing, adaptive control, inference, planning, reasoning, robotics, time series prediction, and DNA sequencing. Our proposed network shows recognition rate (100%). The proposed topology doesnpsilat need a long time for training so it is suitable for on line recognition. It is suitable for configurable hardware implementation and it can be used when there no enough data for training.