English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Plus monthly

Global: Zendy Tools annual

Global: Zendy Tools monthly

Global: Zendy Free Plan

Keyword spotting (KWS) has become an indispensable part of many intelligentdevices surrounding us, as audio is one of the most efficient ways ofinteracting with these devices. The accuracy and performance of KWS solutionshave been the main focus of the researchers, and thanks to deep learning,substantial progress has been made in this domain. However, as the use of KWSspreads into IoT devices, energy efficiency becomes a very critical requirementbesides the performance. We believe KWS solutions that would seek poweroptimization both in the hardware and the neural network (NN) modelarchitecture are advantageous over many solutions in the literature wheremostly the architecture side of the problem is considered. In this work, wedesigned an optimized KWS CNN model by considering end-to-end energy efficiencyfor the deployment at MAX78000, an ultra-low-power CNN accelerator. With thecombined hardware and model optimization approach, we achieve 96.3\% accuracyfor 12 classes while only consuming 251 uJ per inference. We compare ourresults with other small-footprint neural network-based KWS solutions in theliterature. Additionally, we share the energy consumption of our model inpower-optimized ARM Cortex-M4F to depict the effectiveness of the chosenhardware for the sake of clarity.

Ultra-Low Power Keyword Spotting at the Edge