VOICE RECOGNITION BASED MOBILE MANIPULATOR FOR HOME AUTOMATION

Abstract

Advances in naturalaspeech processing have enabledasignificantly moreapowerful speech processingasystems, primarily due totthe use of machine learning techniques. To integrate cloud-based speech recognition system for humanarobot interaction, an interface for voice recognition was going to be developed so that the mobile manipulator can capture those commands by the user and provide feedback or response as per the indent of the user. The main contribution of this work is the design and implementation of a software interface to recognize commands and subsequent conversion through cloud-based speech processing Machine readableccode. Requirementsafor the evaluation offdifferent cloud-services such as Google speech recognition system amazon Alexa and furthermore for the control of robots are determined. But before that simulation is to be done on the python program via Jupyter notebook on implementation of speech recognition to control the mobile manipulator. The simulation is carried out using a differential drive mobile manipulator in this work. The simulation is run in PyCharm IDE, which displays an animation of the differential drive manipulator. First, the simulation is run without using voice by driving the manipulator wheels with a laptop keyboard in an open space (with no obstructions), and then the simulation is run with speech. Extending it even further, replicating voice-based mobile manipulating in a home context for obstacle avoidance.