A particle swarm optimization algorithm for Mandarin speech recognition

In order to remove the human factors improve efficiency and save the cost, National Mandarin Proficiency Test (NMPT) based on computer-aided is expected urgently. The feature parameter extraction is the most important part in the objective assessment of the Mardarin. In this paper we focus on the optimization of artificial neural network by PSO for Mandarin speech recognition. A framework of the Mandarin speech recognition is presented. The PSO algorithm is utilized to adjust the connection weights of the selected ANN topology. Experimental results show the assessment results which obtained by PSO-based ANN method were closer to the ideal output value and more precise.