On-Device Domain Learning for Keyword Spotting on Low-Power Extreme Edge Embedded Systems


METADATA ONLY
Loading...

Date

2024

Publication Type

Conference Paper

ETH Bibliography

yes

Citations

Altmetric
METADATA ONLY

Data

Rights / License

Abstract

Keyword spotting accuracy degrades when neural networks are exposed to noisy environments. On-site adaptation to previously unseen noise is crucial to recovering accuracy loss, and on-device learning is required to ensure that the adaptation process happens entirely on the edge device. In this work, we propose a fully on-device domain adaptation system achieving up to 14% accuracy gains over already-robust keyword spotting models. We enable on-device learning with less than 10 kB of memory, using only 100 labeled utterances to recover 5% accuracy after adapting to the complex speech noise. We demonstrate that domain adaptation can be achieved on ultra-low-power microcontrollers with as little as 806mJ in only 14 s on always-on, battery-operated devices.

Publication status

published

Editor

Book title

2024 IEEE 6th International Conference on AI Circuits and Systems (AICAS)

Journal / series

Volume

Pages / Article No.

6 - 10

Publisher

IEEE

Event

6th IEEE International Conference on AI Circuits and Systems (AICAS 2024)

Edition / version

Methods

Software

Geographic location

Date collected

Date created

Subject

On-Device Learning; Domain Adaptation; Low-Power Microcontrollers; Extreme Edge; TinyML; Noise Robustness; Keyword Spotting

Organisational unit

03996 - Benini, Luca / Benini, Luca check_circle

Notes

Funding

Related publications and datasets