作者：宋亚男[1]林锡海[1]徐荣华[1]宋子寅[1]Song Yanan, Lin Xihai, Xu Ronghua, Song Ziyin (College of Automation, Guangdong University of Technology, Guangzhou 510006, China)
出处：《实验技术与管理》第2期36-38,共3页Experimental Technology and Management
基金：广东省实验室管理专业委员会研究基金项目(033);广东省大学生创新实验项目(1184510031);广东省大学生创新创业训练计划项目(1184512038);广东工业大学合生珠江大学生实验创新项目(HSZJ019);广东工业大学高教研究基金项目(Y09)
摘要：基于Matlab设计了以动态时间规整(dynamic rime warping, DTW)算法和Mel频率倒谱系数(Mel frequency cepstrum coefficients, MFCC)参数提取算法为基础的孤立词自动语音识别系统；然后结合机器人语音识别的需求，基于凌阳SPCE061A设计了语音识别应用系统。结合上述两部分工作，设计、实现了机器人语音识别系统演示实验和机器人语音识别半开放实验，实现效果达到预期实验设计目标。The isolated-words speech recognition system is implemented based on dynamic time warping algo- rithm and Mel frequency cepstrum coefficients parameters extraction in Matlab. And then robot speech recog- nition is implemented based on Lingyang SPCE061A considering application requirements of robot. Based on work above, demonstrating experiment and experiment with open algorithm of robot speech recognition are de- signed. Results show that experiments designed meet the design aims.
