摘要： 阐述语音人机交互手段的必要性及意义，提出一种基于语音控制的机器狗系统解决方案。该系统利用HTK和Julius构建大词汇量连续的语音识别平台，采用语音识别技术提取语音命令用于机器狗控制，并结合有限状态机的控制理论设计机器狗控制系统模型。测试结果表明，该系统能较好地识别用户语音命令，控制系统模型能有效完成机器狗在不同状态下的控制任务。

关键词: 语音识别, 有限状态机, 多智能体, 机器人, 控制模型

Abstract: This article states the necessity of human-computer interaction on the basis of speech and proposes the implementation of speech control robot dog system. HTK and Julius are applied to build a large vocabulary continuous speech recognition platform. Besides, speech recognition technique is used to extract speech command to control the robot dog. A Finite State Machine(FSM) model is adopted to design the control system of the robot dog. Experimental results show this system does good in recognizing user’s speech command. The control system can complete the control tasks under different conditions effectively.

Key words: speech recognition, Finite State Machine(FSM), multi-agent, robot, control model

中图分类号: 

• N945