Specification of a smart-analysis system of sound events for smart environments

Abstract

Purpose

In this work, we proposed a smart-analysis system of sound events for smart/intelligent environments based on an autonomic cycle of data analysis tasks. Design/methodology/approach

We propose an autonomic cycle of data analysis tasks. An autonomic cycle of data analysis tasks is a set of data analysis tasks that supervise and control a process anonymously, which are based on knowledge models (of prediction, recognition, etc.), interacting with each other to reach a common goal. Each task has a different function in the cycle: observation of the process, or its analysis, or decision-making. Findings

This work presents the autonomic cycle. With its components, this autonomic cycle detects sound information using a taxonomic model of the sound events to analyze them and give a recommendation about the context. The taxonomic model is a hierarchical pattern that considers different aspects to recognize the sound events. This work defines the architecture of this autonomic cycle, specifies its machine-learning-based analysis tasks and evaluates its capabilities of reasoning, adaptation and communication in case studies. Research limitations/implications

It is important to work in the future on the improvement of the accuracy of the system by implementing neural networks or more sophisticated techniques. To take the implemented autonomic cycle to a higher level, it could use parallel function management. The automation also needs improvement. In addition to that, future works are going to be directed not only to sound events but also to include emotion recognition and its relation with sound events happening simultaneously. Originality/value

The main contributions of this paper are as follows: the detailed description of the intelligent sound analysis (ISA) autonomic cycle for the smart sound analysis of sound events (SAS-SE) in an intelligent environment (IE); the specification of the machine-learning-based analysis tasks of ISA for the smart sound analysis and the development of a case study that settles the use of the system in different IEs.