摘要
本文基于NAO机器人设计了一种双模态(文本和表情)融合的情感计算系统,以提高机器人对情感识别的准确率,赋予人形机器人NAO“情感”。利用OpenCV开源库和VGG神经网络识别出表情;利用百度语音识别API和BERT神经网络识别出话语内容的情感,并利用判决层融合技术得到最终的情感。经测试,该系统达到了较为和谐的人机交互效果。
In this paper,based on the NAO robot,a dual-mode(text and expression)fusion emotion computing system is designed to improve the accuracy of emotion recognition and give the robot NAO"emotion".We use OpenCV open source library and VGG neural network to recognize the expression,Baidu speech recognition API and BERT neural network to recognize the emotion of speech content,and finally we use decision level fusion technology to get the final emotion.After testing,the system in this paper achieves a more harmonious human-computer interaction effect.
作者
龙林
金永军
程帅杰
廖湘衡
Long Lin;Jin Yongjun;Cheng Shuaijie;Liao Xiangheng(School of Computer Science,South-central Minzu University,Wuhan 430074,China)
出处
《单片机与嵌入式系统应用》
2020年第11期21-23,共3页
Microcontrollers & Embedded Systems
关键词
人形机器人
神经网络
双模态
情感计算
humanoid robot
neural networks
dual-mode
emotion computing
