Smartphone-Based Online and Offline Speech Recognition System for ROS-Based Robots

Speech recognition is one of the fundamental requirements for fully autonomous robotic systems nowadays. The objective of the presented work is to offer a smartphone based speech recognition system for ROS (Robot Operating System) based autonomous robotic systems. The proposed recognition process consists of three steps, namely acquisition, preprocessing, and result extraction. In the contribution we also present a comparison by applying the proposed recognition process using Online and Offline speech servers. The Online method uses Google speech server whereas the Offline uses Simon speech server. The speech recognition and the robotic controlling system work under a server-client architecture. For the evaluation of the Online, we used an Android smartphone, a Pioneer-3DX robot, and Google speech server. For the Offline recognition evaluation, an Android smartphone, a Pioneer-3AT and a standalone Simon speech server are used. The applications running on the smartphone act like clients and communicate with both speech server and robotic system. The contribution provided in this paper is two fold: One using smartphone with Offline speech recognition server Simon, and on the other hand, utilizing Simon's voice recognition feature with robotic navigation on ROS platform.