Automatic Environmental Sound Recognition: Performance versus Computational Cost
This work addresses the need for efficient sound recognition in IoT devices with limited computing power, though it is incremental as it compares existing methods rather than introducing new ones.
The study compared Automatic Environmental Sound Recognition algorithms to determine which provides the best classification performance relative to computational cost, finding that Deep Neural Networks offer the highest accuracy-to-cost ratio, Gaussian Mixture Models provide reasonable accuracy at low cost, and Support Vector Machines offer a middle-ground compromise.
In the context of the Internet of Things (IoT), sound sensing applications are required to run on embedded platforms where notions of product pricing and form factor impose hard constraints on the available computing power. Whereas Automatic Environmental Sound Recognition (AESR) algorithms are most often developed with limited consideration for computational cost, this article seeks which AESR algorithm can make the most of a limited amount of computing power by comparing the sound classification performance em as a function of its computational cost. Results suggest that Deep Neural Networks yield the best ratio of sound classification accuracy across a range of computational costs, while Gaussian Mixture Models offer a reasonable accuracy at a consistently small cost, and Support Vector Machines stand between both in terms of compromise between accuracy and computational cost.