摘要
通过对基于隐马尔科夫模型(Hidden Markov model,HMM)的语音识别算法进行研究,将HMM模型算法的基本思想应用到机器人语音识别系统中,以Sphinx为测试平台,对机器人的控制命令语音信号进行训练得到语言模型和声学模型,利用训练得到的语言模型和声学模型构建一个机器人控制命令语音识别系统,实验测试结果表明,该系统平均错词率为7.1%,具有良好的识别效果,在小词汇量汉语语音识别中具有较高的识别率。
This paper do research about the speech recognition algorithm based on Hidden Markov model (HMM),the basic concept of HMM algorithm is applied to the robot speech recognition system.Taking Sphinx as test platform, the speech signal of the robot is trained to get the language model and acoustic model.Using language model and acoustic model of training to build a speech recognition system of robot control command.The experiment results show that the word error rate of the system is 2.3%, it has good recognition effect, especially shows higher recognition rate in small vocabulary chinese speech recognition.
作者
袁翔
YUAN Xiang (Faculty of Information Engineering, Jiangxi University of Science and Technology, Ganzhou 341000, China)
出处
《电脑知识与技术》
2017年第3期154-155,161,共3页
Computer Knowledge and Technology