keyword spotting (KWS)

  • No Comments

A 0.4-V 988-nW Tiny Footprint Time-Domain Audio Feature Extraction ASIC for Keyword Spotting Using Injection-Locked Oscillators

A 0.4-V 988-nW Tiny Footprint Time-Domain Audio Feature Extraction ASIC for Keyword Spotting Using Injection-Locked Oscillators 150 150
Author(s): Ali Mostafa, Emmanuel Hardy, Franck Badets

Abstract:

This work presents an injection-locked oscillator (ILO)-based feature extraction (FEx) system. It combines voltage- and time-domain signal processing to implement a power-efficient programmable gain amplifier (PGA) and a small-footprint, high-selectivity ILO-based voltage-to-time converter bandpass filter (VTC-BPF) bank and rectifier. The VTC-BPF enables direct analog-to-time conversion, eliminating the need for …

Published in: IEEE Journal of Solid-State Circuits
Page(s): 1 – 13
Year of Publication: 2026
Electronic ISSN: 1558-173X
DOI: 10.1109/JSSC.2026.3681481
Publisher: IEEE

View on IEEE Xplore

  • No Comments

A 7.5-μW 35-Keyword End-to-End Keyword Spotting System With Random Augmented On-Chip Training

A 7.5-μW 35-Keyword End-to-End Keyword Spotting System With Random Augmented On-Chip Training 150 150
Author(s): Hyuk-Jin Lee, Kyunghoon Pyo, Taekwang Jang, Mingoo Seok, SeongHwan Cho

Abstract:

Fully integrated keyword spotting (KWS) systems designed for low-power operation face two major challenges. First, increasing the number of supported keywords significantly raises system complexity and power consumption. Second, most existing systems are not personalized to individual users, as they are trained on data from native English speakers, leading to …

Published in: IEEE Journal of Solid-State Circuits
Page(s): 1 – 10
Year of Publication: 2026
Electronic ISSN: 1558-173X
DOI: 10.1109/JSSC.2026.3660634
Publisher: IEEE

View on IEEE Xplore

  • No Comments

A 475-nW Area-Efficient Programmable Analog Feature Extraction Filterbank for Audio Classification

A 475-nW Area-Efficient Programmable Analog Feature Extraction Filterbank for Audio Classification 150 150
Author(s): Patrice Abbie D. Legaspi, Shubham Yadav, Mark S. Oude Alink, Bram Nauta

Abstract:

Audio classification in edge devices has many applications and can be implemented at varying levels of complexity, typically consisting of a feature extractor followed by a classifier. Such devices are often always-on, constantly listening to their surroundings, and have a small form factor; therefore, they require low-power operation and high …

Published in: IEEE Journal of Solid-State Circuits
Page(s): 1 – 15
Year of Publication: 2026
Electronic ISSN: 1558-173X
DOI: 10.1109/JSSC.2026.3655426
Publisher: IEEE

View on IEEE Xplore

  • No Comments

Chameleon: A Multiplier-Free Temporal Convolutional Network Accelerator for End-to-End Few-Shot and Continual Learning from Sequential Data

Chameleon: A Multiplier-Free Temporal Convolutional Network Accelerator for End-to-End Few-Shot and Continual Learning from Sequential Data 150 150
Author(s): Douwe den Blanken, Charlotte Frenkel

Abstract:

On-device learning at the edge enables low-latency, private personalization with improved long-term robustness and reduced maintenance costs. Yet, achieving scalable, low-power (LP) end-to-end on-chip learning, especially from real-world sequential data with a limited number of examples, is an open challenge. Indeed, accelerators supporting error backpropagation optimize for learning performance at …

Published in: IEEE Journal of Solid-State Circuits
Page(s): 1 – 16
Year of Publication: 2026
Electronic ISSN: 1558-173X
DOI: 10.1109/JSSC.2025.3645640
Publisher: IEEE

View on IEEE Xplore

Search this website