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/

Zendy Plus

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

Zendy Tools

Zendy Open

At the end of Moore's law, new computing paradigms are required to prolongthe battery life of wearable and IoT smart audio devices. Theoretical analysisand physical validation have shown that analog signal processing (ASP) can bemore power-efficient than its digital counterpart in the realm of low-to-mediumsignal-to-noise ratio applications. In addition, ASP allows a direct interfacewith an analog microphone without a power-hungry analog-to-digital converter.Here, we present power-efficient analog acoustic features (PEAF) that arevalidated by fabricated CMOS chips for running audio recognition. Linear,non-linear, and learnable PEAF variants are evaluated on two speech processingtasks that are demanded in many battery-operated devices: wake word detection(WWD) and keyword spotting (KWS). Compared to digital acoustic features, higherpower efficiency with competitive classification accuracy can be obtained. Anovel theoretical framework based on information theory is established toanalyze the information flow in each individual stage of the feature extractionpipeline. The analysis identifies the information bottleneck and helps improvethe KWS accuracy by up to 7%. This work may pave the way to building morepower-efficient smart audio devices with best-in-class inference performance.

PEAF: Learnable Power Efficient Analog Acoustic Features for Audio Recognition