Voice Expression System of Visual Environment for a Guide Dog Robot

摘要

In this paper, a voice expression system of visual environment for a guide dog robot is proposed. This system enables the robot to recognize objects and scenes in front of the robot by CNN (Convolutional Neural Network) and generate captions of the scenes by LSTM (Long-Short-Term Memory) network. Then the robot expresses the recognized visual scene by voice, which is generated by speech synthesis. The guide dog robot can guide the visually impaired person safely to the desired destination, as well as entertain the user by expressing various visual information through voice. The system is composed of object recognition, scene caption, and speech synthesis. The effectiveness of this system is confirmed through experiments conducted with our guide dog robot.