系统采用凌阳SPCE061A 单片机作为语音识别系统的主控芯片。通过硬件电路 设计和软件代码部分成功的设计并实现了一种具有语音识别功能、语音提示(语音合成)及语音回放(语音编码记录)功能的嵌入式语音识别系统。语音识别模型采用(DHMM)离散隐马尔可夫模型,利用Baum-welth 重估算法、前向后向算法、viterbi 算法来完成语音模板的训练和语音识别的任务。 关键词:SPCE061A 单片机;语音识别系统;语音识别模型 Design of Speech Recognition System Base on SPCE061A Yingfeng fengyufen Abstract: The system selects SPCE061A as MCU of speech recognition system. Through the hardware circuit design and the software code, design and realize the embedded speech recognition system which contains functions of speech recognition, voice prompt (speech synthesis) and audio playback (speech coding record). The discrete hidden markov model(DHMM)is adopted,Baum-Welth reestimation algorithm, forward-backward procedure and viterbi algorithm is utilized to train and recognize the speech signal. Keywords: SPCE061A MCU; Speech Recognition System; Speech Recognition Model