OnDA: On-device Channel Pruning for Efficient Personalized Keyword Spotting
Fuente:
arXiv
Salvato in:
| Autori principali: | , , |
|---|---|
| Natura: | Preprint |
| Pubblicazione: |
2026
|
| Soggetti: | |
| Accesso online: | |
| Tags: |
Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
|
| _version_ | 1866910038794174464 |
|---|---|
| author | Risso, Matteo Burrello, Alessio Pagliari, Daniele Jahier |
| author_facet | Risso, Matteo Burrello, Alessio Pagliari, Daniele Jahier |
| contents | Always-on keyword spotting (KWS) demands on-device adaptation to cope with user- and environment-specific distribution shifts under tight latency and energy budgets. This paper proposes, for the first time, coupling weight adaptation (i.e., on-device training) with architectural adaptation, in the form of online structured channel pruning, for personalized on-device KWS. Starting from a state-of-the-art self-learning personalized KWS pipeline, we compare data-agnostic and data-aware pruning criteria applied on in-field pseudo-labelled user data. On the HeySnips and HeySnapdragon datasets, we achieve up to 9.63x model-size compression with respect to unpruned baselines at iso-task performance, measured as the accuracy at 0.5 false alarms per hour. When deploying our adaptation pipeline on a Jetson Orin Nano embedded GPU, we achieve up to 1.52x/1.57x and 1.64x/1.77x latency and energy-consumption improvements during online training/inference compared to weights-only adaptation. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2603_02247 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | OnDA: On-device Channel Pruning for Efficient Personalized Keyword Spotting Risso, Matteo Burrello, Alessio Pagliari, Daniele Jahier Audio and Speech Processing Machine Learning Sound Always-on keyword spotting (KWS) demands on-device adaptation to cope with user- and environment-specific distribution shifts under tight latency and energy budgets. This paper proposes, for the first time, coupling weight adaptation (i.e., on-device training) with architectural adaptation, in the form of online structured channel pruning, for personalized on-device KWS. Starting from a state-of-the-art self-learning personalized KWS pipeline, we compare data-agnostic and data-aware pruning criteria applied on in-field pseudo-labelled user data. On the HeySnips and HeySnapdragon datasets, we achieve up to 9.63x model-size compression with respect to unpruned baselines at iso-task performance, measured as the accuracy at 0.5 false alarms per hour. When deploying our adaptation pipeline on a Jetson Orin Nano embedded GPU, we achieve up to 1.52x/1.57x and 1.64x/1.77x latency and energy-consumption improvements during online training/inference compared to weights-only adaptation. |
| title | OnDA: On-device Channel Pruning for Efficient Personalized Keyword Spotting |
| topic | Audio and Speech Processing Machine Learning Sound |
| url | https://arxiv.org/abs/2603.02247 |