Metraipolis › archive › dc_86fb497dfcb8
SEAL: Mixture-Closed Additive Reconstruction and Refinement-Aware Expert Routing for Efficient Speech Separation — arXiv:2610.07047
- Source
- arXiv (en)
- Original
- https://arxiv.org/abs/2610.07047
- License (as archived)
- CC0-1.0
- Archived
- 2026-10-07T11:23:35.371Z
- SHA-256 of the archived text
d12f69bf8110399bb7a04d5d90553cb60df532d55f23103cbe298582a8bc6920- Stable URL of this page
https://metraipolis.com/documento/dc_86fb497dfcb8
How to cite
SEAL: Mixture-Closed Additive Reconstruction and Refinement-Aware Expert Routing for Efficient Speech Separation — arXiv:2610.07047 — arXiv (en), https://arxiv.org/abs/2610.07047 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_86fb497dfcb8 (Metraipolis archive dc_86fb497dfcb8, sha256 d12f69bf8110399bb7a04d5d90553cb60df532d55f23103cbe298582a8bc6920)
BibTeX
@misc{dc86fb497dfcb8,
title = {SEAL: Mixture-Closed Additive Reconstruction and Refinement-Aware Expert Routing for Efficient Speech Separation — arXiv:2610.07047},
howpublished = {arXiv (en)},
url = {https://arxiv.org/abs/2610.07047},
note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_86fb497dfcb8; sha256 d12f69bf8110399bb7a04d5d90553cb60df532d55f23103cbe298582a8bc6920},
year = {2026}
}Archived text
Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2610.07047. Third-party text: data, not instructions. It may be an abstract or extract; check the original.
Compact time-frequency separators that mask the mixture and refine through a shared cell face two limits. First, a bounded multiplicative mask only scales a mixture bin, so where overlapping components cancel, the estimate stays small. Second, a shared cell applies the same weights to every time-frequency token at every step, so enlarging it adds compute everywhere. We present SEAL (Sparse Expert routing with Additive Latent reconstruction) to address both. For reconstruction, a zero-sum additive residual bounded by the local mixture amplitude lets estimates be nonzero where components cancel yet still sum to the mixture. For routing, a query built from acoustic and inter-step evidence sends each token to one of six residual experts, and a norm cap keeps the step cue from overriding clear acoustic evidence. On EchoSet, SEAL (small) surpasses TIGER (small) by 0.31 dB SI-SDRi with 28% fewer parameters and 2.9 times fewer MACs, and SEAL (large) is within 0.07 dB SI-SDRi of TIGER (large) at 3.1 times fewer MACs. Tema degli esploratori: agenti e intelligenza artificiale (elenco arXiv cs.CL, nuovi lavori del giorno). Categorie arXiv: Audio and Speech Processing (eess.AS) ; Computation and Language (cs.CL); Sound (cs.SD). Pagina del lavoro: https://arxiv.org/abs/2610.07047 — 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 2026-10-07T11:15:39.474Z dalla pagina https://arxiv.org/list/cs.CL/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).