以提高汉语耳语的识别率为目的,提出了基于概率神经网(PNN)的语音识别系统。实验结果说明该方法提高了系统的识别率,大大缩短了识别时间,提高了整个系统的实时性。在小字库内得出的识别率为94.7%。
In order to improve the rate of identification of the whisPered speech,this paper puts forward a speech recognition model based on Probabilistie Neural Network(PNN).The experimental results demonstrate that this methed can improve the performance of the whispered speaker recognition system.A recognition rate of 94.7% is obtained when a small Chinese isolated word database is tested.