Adaptive Knowledge Distillation for Device-Directed Speech Detection
Fuente:
arXiv
Saved in:
| Main Authors: | , , , , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2025
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866908478617944064 |
|---|---|
| author | Chi, Hyung Gun Pesce, Florian Chang, Wonil Rudovic, Oggi Argueta, Arturo Braun, Stefan Garg, Vineet Abdelaziz, Ahmed Hussen |
| author_facet | Chi, Hyung Gun Pesce, Florian Chang, Wonil Rudovic, Oggi Argueta, Arturo Braun, Stefan Garg, Vineet Abdelaziz, Ahmed Hussen |
| contents | Device-directed speech detection (DDSD) is a binary classification task that separates the user's queries to a voice assistant (VA) from background speech or side conversations. This is important for achieving naturalistic user experience. To this end, we propose knowledge distillation (KD) to enhance DDSD accuracy while ensuring efficient deployment. Specifically, we introduce a novel adaptive KD method that transfers knowledge from general representations of an ASR large pre-trained acoustic encoder (teacher). We apply task-specific adapters, on top of the (frozen) teacher encoder, trained jointly with the student model on DDSD. We demonstrate that the proposed adaptive KD outperforms the student model without distillation in the keyword and keyword-free (follow-up) invocations, with an improvement of +26% and +19% in terms of Equal Error Rate, respectively. We also show that this approach generalizes across the transformer and conformer-based model architectures. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2508_02801 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | Adaptive Knowledge Distillation for Device-Directed Speech Detection Chi, Hyung Gun Pesce, Florian Chang, Wonil Rudovic, Oggi Argueta, Arturo Braun, Stefan Garg, Vineet Abdelaziz, Ahmed Hussen Sound Artificial Intelligence Audio and Speech Processing Device-directed speech detection (DDSD) is a binary classification task that separates the user's queries to a voice assistant (VA) from background speech or side conversations. This is important for achieving naturalistic user experience. To this end, we propose knowledge distillation (KD) to enhance DDSD accuracy while ensuring efficient deployment. Specifically, we introduce a novel adaptive KD method that transfers knowledge from general representations of an ASR large pre-trained acoustic encoder (teacher). We apply task-specific adapters, on top of the (frozen) teacher encoder, trained jointly with the student model on DDSD. We demonstrate that the proposed adaptive KD outperforms the student model without distillation in the keyword and keyword-free (follow-up) invocations, with an improvement of +26% and +19% in terms of Equal Error Rate, respectively. We also show that this approach generalizes across the transformer and conformer-based model architectures. |
| title | Adaptive Knowledge Distillation for Device-Directed Speech Detection |
| topic | Sound Artificial Intelligence Audio and Speech Processing |
| url | https://arxiv.org/abs/2508.02801 |