Papers › Position-invariant Fine-tuning of Speech Enhancement Models with Self-supervised...

Position-invariant Fine-tuning of Speech Enhancement Models with Self-supervised Speech Representations

28 Jan 2026arXiv:2601.21084added by Syntology

Amit Meghanani, Thomas Hain

Title, abstract, authors and date from arXiv's metadata (CC0); this paper is not in the Papers with Code archive (frozen 2025-07-28).

Integrating front-end speech enhancement (SE) models with self-supervised learning (SSL)-based speech models is effective for downstream tasks in noisy conditions. SE models are commonly fine-tuned using SSL representations with mean squared error (MSE) loss between enhanced and clean speech. However, MSE is prone to exploiting positional embeddings in SSL models, allowing the objective to be minimised through positional correlations instead of content-related information. This work frames the problem as a general limitation of self-supervised representation fine-tuning and investigates it through representation-guided SE. Two strategies are considered: (1) zero-padding, previously explored in SSL pre-training but here examined in the fine-tuning setting, and (2) speed perturbations with a soft-DTW loss. Experiments show that the soft-DTW-based approach achieves faster convergence and improved downstream performance, underscoring the importance of position-invariant fine-tuning in SSL-based speech modelling.

PaperPDF

In Syntology Open this paper in Syntology's Atlas, the map of the papers in Syntology's graph and their citations.

Code

facebookresearch/denoiser found in paper text by Syntology report
s3prl/s3prl found in paper text by Syntology report

Reachability, where shown, is from one Syntology probe window (2026-09-16 to 2026-09-18); repositories not probed show nothing. GitHub stars are not tracked.

Code Syntology ran Syntology

Syntology holds the repository link but has not harvested or run code from it.

Results from the paper

The Papers with Code archive ends with its 2025-07-28 snapshot. This paper's arXiv identifier, 2601.21084, was issued in January 2026, after that date, so the archive has no leaderboard rows for it.

Report a problem or propose a change · a person checks every report against the paper or source before anything changes; decisions are listed on /corrections