On-Device Domain Learning for Keyword Spotting on Low-Power Extreme Edge Embedded Systems
Authors:
- 22 April 2024
- conference paper
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 6-10
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 806 mJ in only 14 s on always-on, battery-operated devices.Keywords
All Related Versions
Funding Information
- National Science Foundation
This publication has 17 references indexed in Scilit:
- Accelerating RNN-Based Speech Enhancement on a Multi-core MCU with Mixed FP16-INT8 Post-training Quantization2023
- PULP-TrainLib: Enabling On-Device Training for RISC-V Multi-core MCUs Through Performance-Driven Autotuning2022
- Towards On-device Domain Adaptation for Noise-Robust Keyword Spotting2022
- Convmixer: Feature Interactive Convolution with Curriculum Learning for Small Footprint and Noisy Far-Field Keyword Spotting2022
- Deep Spoken Keyword Spotting: An OverviewIEEE Access, 2021
- TinyOL: TinyML with Online-Learning on Microcontrollers2021
- A Depthwise Separable Convolution Neural Network for Small-footprint Keyword Spotting Using Approximate MAC Unit and Streaming Convolution Reuse2019
- Deep Learning With Edge Computing: A ReviewProceedings of the IEEE, 2019
- Hotword Cleaner: Dual-microphone Adaptive Noise Cancellation with Deferred Filter Coefficients for Robust Keyword Spotting2019
- Supervised Noise Reduction for Multichannel Keyword Spotting2018