Robot Assisting Water Serving to Disabilities by Voice Control

Abstract

ROS is an open-source robot operating system. In this paper, we use ROS to control Conbe robot arm. By introducing the YOLACT real-time instance segmentation, we trained our own model for Object Detection. Secondly, the Speech-Recognition system is established through Deep speech and Mozilla Text-To-Speech with Tacotron2 DDC model. Deep speech is an end-to-end speech system, where deep learning supersedes these processing stages. Combined with a language model, this approach achieves higher performance than traditional methods on hard speech recognition tasks while also being much simpler. In this way, we create an artificial intelligence, which accomplished a simple conversation with people. And the voice control system is established based on Speech-Recognition system. In the experiment, we successfully control the robot arm move positions and do water serving for disabilities by voice command. With this research, voice control robot arm can be apply in the life support area, it will be more convenient for disabilities in daily life.