In order to improve the speech interaction performance of intelligent question answering robot,a speech interaction control method based on speech recognition is proposed.The Teager energy of the speech command signal issued by the user to the intelligent question answering robot is used to extract the speech feature of the speech command signal.Based on the speech feature of the extracted speech command signal,the acoustic model of the intelligent question answering robot is constructed u-sing the hidden Markov model,and the speech command received by the intelligent question answering robot is recognized using the acoustic model.The speech command recognition result is taken as the input of the least square support vector machine,and the response result of the intelligent question answering robot is taken as the output of the least square support vector ma-chine to complete the speech interaction control of the intelligent question answering robot.The experimental results show that the phoneme error rate and word error rate of speech commands recognized by this method are lower than 0.7％,and good re-sults of intelligent question answering robot speech interaction can be obtained.
关键词
语音识别/智能问答机器人/语音交互/控制方法/隐马尔科夫模型
Key words
speech recognition/intelligent question answering robot/speech interaction/control method/hidden Markov model
维普
万方数据