Saved in:
| Main Authors: | , , , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2025
|
| Subjects: | |
| Online Access: | https://arxiv.org/abs/2507.09768 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866914368358187008 |
|---|---|
| author | Olsen, Kenny Falkær Østergaard, Mads Ulbæk, Karl Nielsen, Søren Føns Lindrup, Rasmus Malik Høegh Jensen, Bjørn Sand Mørup, Morten |
| author_facet | Olsen, Kenny Falkær Østergaard, Mads Ulbæk, Karl Nielsen, Søren Føns Lindrup, Rasmus Malik Høegh Jensen, Bjørn Sand Mørup, Morten |
| contents | In recent years, deep learning-based single-channel speech separation has improved considerably, in large part driven by increasingly compute- and parameter-efficient neural network architectures. Most such architectures are, however, designed with a fixed compute and parameter budget and consequently cannot scale to varying compute demands or resources, which limits their use in embedded and heterogeneous devices such as mobile phones and hearables. To enable such use-cases we design a neural network architecture for speech separation and enhancement capable of early-exit, and we propose an uncertainty-aware probabilistic framework to jointly model the clean speech signal and error variance which we use to derive probabilistic early-exit conditions in terms of desired signal-to-noise ratios. We evaluate our methods on both speech separation and enhancement tasks where we demonstrate that early-exit capabilities can be introduced without compromising reconstruction, and that when trained on variable-length audio our early-exit conditions are well-calibrated and lead to considerable compute savings when used to dynamically scale compute at test time while remaining directly interpretable. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2507_09768 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | Knowing When to Quit: Probabilistic Early Exits for Speech Separation Olsen, Kenny Falkær Østergaard, Mads Ulbæk, Karl Nielsen, Søren Føns Lindrup, Rasmus Malik Høegh Jensen, Bjørn Sand Mørup, Morten Machine Learning Sound Audio and Speech Processing In recent years, deep learning-based single-channel speech separation has improved considerably, in large part driven by increasingly compute- and parameter-efficient neural network architectures. Most such architectures are, however, designed with a fixed compute and parameter budget and consequently cannot scale to varying compute demands or resources, which limits their use in embedded and heterogeneous devices such as mobile phones and hearables. To enable such use-cases we design a neural network architecture for speech separation and enhancement capable of early-exit, and we propose an uncertainty-aware probabilistic framework to jointly model the clean speech signal and error variance which we use to derive probabilistic early-exit conditions in terms of desired signal-to-noise ratios. We evaluate our methods on both speech separation and enhancement tasks where we demonstrate that early-exit capabilities can be introduced without compromising reconstruction, and that when trained on variable-length audio our early-exit conditions are well-calibrated and lead to considerable compute savings when used to dynamically scale compute at test time while remaining directly interpretable. |
| title | Knowing When to Quit: Probabilistic Early Exits for Speech Separation |
| topic | Machine Learning Sound Audio and Speech Processing |
| url | https://arxiv.org/abs/2507.09768 |