A probabilistic neural network for spatiotemporal pattern recognition

A spatiotemporal probabilistic neural network (SPNN) is proposed for spatiotemporal pattern recognition. Speaker-independent isolated Mandarin digit recognition is used as an example of spatiotemporal pattern recognition. The simulation result shows that SPNN can work quite well for spatiotemporal pattern recognition. The simplicity and effectiveness of SPNN make it suitable for the applications of speech recognition, radar, sonar, etc.<>