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) is a crucial function enabling the interaction withthe many ubiquitous smart devices in our surroundings, either activating themthrough wake-word or directly as a human-computer interface. For manyapplications, KWS is the entry point for our interactions with the device and,thus, an always-on workload. Many smart devices are mobile and their batterylifetime is heavily impacted by continuously running services. KWS and similaralways-on services are thus the focus when optimizing the overall powerconsumption. This work addresses KWS energy-efficiency on low-costmicrocontroller units (MCUs). We combine analog binary feature extraction withbinary neural networks. By replacing the digital preprocessing with theproposed analog front-end, we show that the energy required for dataacquisition and preprocessing can be reduced by 29x, cutting its share from adominating 85% to a mere 16% of the overall energy consumption for ourreference KWS application. Experimental evaluations on the Speech CommandsDataset show that the proposed system outperforms state-of-the-art accuracy andenergy efficiency, respectively, by 1% and 4.3x on a 10-class dataset whileproviding a compelling accuracy-energy trade-off including a 2% accuracy dropfor a 71x energy reduction.

Sub-mW Keyword Spotting on an MCU: Analog Binary Feature Extraction and Binary Neural Networks