A cloud middleware enabling natural speech analysis for IoT policy enforcement in smart home environments


In this paper, we present a solution for enhancing the IoT-enabled smart home experience by means of natural speech analysis in lieu of specific voice commands. Our proposed cloud middleware detects phrases and analyzes the semantic and syntactic makeup of the phrases to generate appropriate tags to trigger an action for provisioned objects with binary power choices (e.g. lights, fans, locks, etc.). It then sends these tags to our policy-based controller, CoMPES, for execution. Thus, the novelty of this paper lies in natural language processing for IoT policy selection and enforcement. We applied a method for measuring the similarity between short phrases based on synonyms and word order information to determine if a spoken phrase can be mapped to a predefined provisioned phrase with a minimum threshold of 75% similarity. Our experimental results demonstrate the viability of the proposed system for real-time IoT policy enforcement.


Computer Science

Document Type

Conference Proceeding




Audio, Cloud middleware, Policy, Semantic analysis, Smart home, Syntactic analysis

Publication Date


Journal Title

Proceedings - 2018 IEEE International Congress on Internet of Things, ICIOT 2018 - Part of the 2018 IEEE World Congress on Services