Abstract
Smart home is a very hot development area in which voice-based control devices are receiving special attention from major technology companies and researchers. Despite many studies on this problem in the world, there has not been a formal study for the Vietnamese language. In addition, many studies did not offer a solution that can be expanded easily in the future. This paper provides a speech collection and processing software and shares a dataset of speech commands is labeled and organized to the language research community. This study also designs and evaluates Recurrent Neural Networks to apply it to the data collected. The average recognition accuracy on the set of 15 commands for controlling smart home devices is 98.19%. Finally, the paper presents the implementation and performance evaluation of machine learning model on a Raspberry PI-based intelligent home control unit.
Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
