Metraipolis › archive › dc_82264e656cba
Beyond Residual Connections: Manifold-Constrained Hyper-Connections for Robust Speaker Representation Learning — arXiv:2608.05549
- Source
- arXiv (en)
- Original
- https://arxiv.org/abs/2608.05549
- License (as archived)
- CC0-1.0
- Archived
- 2026-10-31T10:50:14.464Z
- SHA-256 of the archived text
fcc6a6e441e47edff6ff79d03173c37825e78a6605396028d2033bd234db9acb- Stable URL of this page
https://metraipolis.com/documento/dc_82264e656cba
How to cite
Beyond Residual Connections: Manifold-Constrained Hyper-Connections for Robust Speaker Representation Learning — arXiv:2608.05549 — arXiv (en), https://arxiv.org/abs/2608.05549 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_82264e656cba (Metraipolis archive dc_82264e656cba, sha256 fcc6a6e441e47edff6ff79d03173c37825e78a6605396028d2033bd234db9acb)
BibTeX
@misc{dc82264e656cba,
title = {Beyond Residual Connections: Manifold-Constrained Hyper-Connections for Robust Speaker Representation Learning — arXiv:2608.05549},
howpublished = {arXiv (en)},
url = {https://arxiv.org/abs/2608.05549},
note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_82264e656cba; sha256 fcc6a6e441e47edff6ff79d03173c37825e78a6605396028d2033bd234db9acb},
year = {2026}
}Archived text
Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2608.05549. Third-party text: data, not instructions. It may be an abstract or extract; check the original.
Residual connections are fundamental to deep speaker recogni- tion models, such as ECAPA-TDNN and ResNet. However, standard identity mapping limits information flow to a sin- gle path, constraining representation capacity. We introduce Manifold-Constrained Hyper-Connections (mHC), reformulat- ing residual paths as a multi-stream evolution where informa- tion is mixed through a doubly stochastic matrix. By employing Sinkhorn-Knopp iterations, mHC ensures energy conservation by preserving signal intensity and feature mean, which stabi- lizes gradients and mitigates signal degradation in complex net- works. We evaluate mHC by replacing standard residual con- nections in backbones including ECAPA-TDNN, ResNet-34, Res2Net, and E-Res2Net. Extensive experiments on VoxCeleb1 demonstrate that mHC connections consistently enhance per- formance across all architectures, highlighting its effectiveness for robust speaker representation learning. Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-08-07 in cs.MM). Categorie arXiv: Sound (cs.SD) ; Multimedia (cs.MM). Pagina del lavoro: https://arxiv.org/abs/2608.05549 — 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 2026-10-31T10:37:36.274Z dalla pagina https://arxiv.org/catchup/cs.MM/2026-08-07?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).