Summary
Audio bandwidth extension aims to fill in missing high-frequency sounds to make recordings clearer and more natural. The authors found that traditional methods often lose rare but important high-frequency bursts, making the audio feel smoothed out or dull. They created a new approach called UDSS-BWE that uses ideas from decision science and uncertainty analysis to better detect and reproduce these subtle high-frequency details. Their method uses an advanced neural network design and special checks to avoid making the audio artificially boosted or noisy. Tests showed their approach produces more realistic audio with fewer model parameters compared to previous methods.
What this means in practice
- •For audio engineers: Improve audio restoration software to better reconstruct natural high-frequency sounds without adding noise or artifacts.
- •For speech enhancement developers: Create more perceptually accurate speech bandwidth extension modules that capture rare transient sounds in noisy environments.
Authors
Tarikul Islam Tamiti, Sajid Fardin Dipto, David Vergano, Luke Baja-Ricketts, Anomadarshi Barua
Abstract
Bandwidth extension (BWE) is fundamentally localized: the most perceptual distortions are not average-case distortions, but rare high-frequency (HF) transients that standard, risk-neutral objectives tend to smooth away. To close this gap, we seek solutions in the risk-sensitive and uncertainty-aware decision science rules and present UDSS-BWE, which introduces five decision-science and uncertainty-aware discriminators: CVaRD (does tail pooling to amplify HF artifacts), CCD (a primal-dual augmented Lagrangian to prevent HF overboost), MCUD (a learnable utility over spectral flatness/ centroid/ rolloff), EDD (captures epistemic uncertainty), and DROD (captures entropic KL-DRO aggregation). UDSS-BWE is also designed as a complex valued adversarial BWE framework that uses Swin-based generators, a lightweight dual-stream shifted-window backbone, to capture local and long-range structure efficiently, while learnable lattice coupling provides controlled cross-stream exchange. UDSS-BWE is optimized extensively and achieves better perceptual quality with 3.89x fewer parameters (72M vs.18.5M) over two English and French datasets under clean and noisy conditions. To the best of our knowledge, this work shows how multi disciplinary decision-science-inspired and uncertainty theories can be successfully used to design efficient discriminators for producing more nuanced audios, establishing a new baseline in the BWE task.