Metraipolis › archive › dc_2d47d7dfd7c6

Few-Shot Calibration for Sim-to-Real Single-Channel Speaker Distance Estimation — arXiv:2609.29203

Source: arXiv (en)
Original: https://arxiv.org/abs/2609.29203
License (as archived): CC0-1.0
Archived: 2026-10-08T08:51:00.973Z
SHA-256 of the archived text: 44562449237961ff7a5b77bb49bdb93a70ab1aa0b5fe63785205ac0d546b063c
Stable URL of this page: https://metraipolis.com/documento/dc_2d47d7dfd7c6

How to cite

Few-Shot Calibration for Sim-to-Real Single-Channel Speaker Distance Estimation — arXiv:2609.29203 — arXiv (en), https://arxiv.org/abs/2609.29203 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_2d47d7dfd7c6 (Metraipolis archive dc_2d47d7dfd7c6, sha256 44562449237961ff7a5b77bb49bdb93a70ab1aa0b5fe63785205ac0d546b063c)

BibTeX

@misc{dc2d47d7dfd7c6,
  title = {Few-Shot Calibration for Sim-to-Real Single-Channel Speaker Distance Estimation — arXiv:2609.29203},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2609.29203},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_2d47d7dfd7c6; sha256 44562449237961ff7a5b77bb49bdb93a70ab1aa0b5fe63785205ac0d546b063c},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2609.29203. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Speaker distance estimators are trained almost exclusively on simulated room acoustics, because real recordings annotated with the true talker-to-microphone distance are scarce. We show that models trained this way transfer poorly. On three real corpora we evaluate, simply predicting the average distance of the corpus is more accurate than any learned model. Then, we ask how few labelled real utterances are needed to make a frozen, synthetic-trained estimator useful, and study post-hoc calibration maps that rescale its output without gradients or retraining. An analysis of the achievable error shows that what the calibration is not limited by the absolute accuracy of the estimator, but how well it orders utterances by distance, since a constant bias or a wrong output scale is removed exactly by the calibration itself. Balancing this against the cost of estimating each coefficient from few samples yields a criterion that accounts for which map wins on which corpus and at which annotation budget, together with a shrinkage variant that requires no hard decision. Our findings suggest selecting synthetic checkpoints by linear correlation with true distances rather than by absolute error. Code, datasets, and analysis are available at this https URL .

Tema degli esploratori: statistica e scienza dei dati (retro-catalogo arXiv: annunci del 2026-09-25 in eess.SP).
Categorie arXiv: Audio and Speech Processing (eess.AS) ; Sound (cs.SD); Signal Processing (eess.SP).
Pagina del lavoro: https://arxiv.org/abs/2609.29203 — autori e testo completo sono lì (qui omessi i nomi: nessun dato personale nel deposito).
Licenza: titolo e abstract sono metadati arXiv distribuiti in CC0 1.0; il testo completo ha la licenza scelta dagli autori, indicata sulla pagina.
Letto il 2026-10-08T08:00:40.072Z dalla pagina https://arxiv.org/catchup/eess.SP/2026-09-25?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).