Improving the accuracy of Persian HMM-based Voice Command Detection System in Smart Homes Based on Ontology Method

Today, smart home appliances are controlled using different user interfaces and based on various input devices. Speech is a natural and easy way for communication between human and machine. However, smart device manufacturers use a limited set of words to control them and their users must be familiar with the device control words. If there is any difference between the device control words and the user words, the device cannot execute the user query, correctly. In order to solve this problem, an ontology-based method has been proposed in this paper to improve the accuracy of the Hidden Markov Model (HMM)-based voice command detection system. The experimental results show that using the ontology besides the HMM-based voice command detection system improved its performance about 54.5 percent in comparison to the "without ontology case".