On-Device Domain Learning for Keyword Spotting on Low-Power Extreme Edge Embedded Systems
Fuente:
arXiv
Saved in:
| Main Authors: | , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2024
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866909138241454080 |
|---|---|
| author | Cioflan, Cristian Cavigelli, Lukas Rusci, Manuele de Prado, Miguel Benini, Luca |
| author_facet | Cioflan, Cristian Cavigelli, Lukas Rusci, Manuele de Prado, Miguel Benini, Luca |
| contents | 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. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2403_10549 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | On-Device Domain Learning for Keyword Spotting on Low-Power Extreme Edge Embedded Systems Cioflan, Cristian Cavigelli, Lukas Rusci, Manuele de Prado, Miguel Benini, Luca Sound Machine Learning Audio and Speech Processing 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. |
| title | On-Device Domain Learning for Keyword Spotting on Low-Power Extreme Edge Embedded Systems |
| topic | Sound Machine Learning Audio and Speech Processing |
| url | https://arxiv.org/abs/2403.10549 |