Metraipolis › archive › dc_163fc7e4fe18

GTCRN-SS: Lightweight Causal Speech Separation with Residual Allocation and Reconstruction Refinement — arXiv:2611.09358

Source
arXiv (en)
Original
https://arxiv.org/abs/2611.09358
License (as archived)
CC0-1.0
Archived
2026-11-09T23:04:21.020Z
SHA-256 of the archived text
7760154203e0bd684e91bf7535dc67296a358de05e91864efee018cbeb4f31ab
Stable URL of this page
https://metraipolis.com/documento/dc_163fc7e4fe18

How to cite

GTCRN-SS: Lightweight Causal Speech Separation with Residual Allocation and Reconstruction Refinement — arXiv:2611.09358 — arXiv (en), https://arxiv.org/abs/2611.09358 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_163fc7e4fe18 (Metraipolis archive dc_163fc7e4fe18, sha256 7760154203e0bd684e91bf7535dc67296a358de05e91864efee018cbeb4f31ab)
BibTeX
@misc{dc163fc7e4fe18,
  title = {GTCRN-SS: Lightweight Causal Speech Separation with Residual Allocation and Reconstruction Refinement — arXiv:2611.09358},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2611.09358},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_163fc7e4fe18; sha256 7760154203e0bd684e91bf7535dc67296a358de05e91864efee018cbeb4f31ab},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2611.09358. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Low-complexity causal speech separation must preserve speaker information and allocate residual mixture energy without relying on future feature frames. We present GTCRN-SS, a two-stage time-frequency separator that adapts the compact GTCRN enhancement backbone to fixed-output speech separation. The model combines internal residual allocation, reconstruction-conditioned refinement, and frame-aligned recurrent-state transfer. Its complex-mask estimator assigns non-target content to an unreturned residual branch, while the complementary speech estimate conditions a second stage whose final masks remain anchored to the original mixture. A lightweight low-rank adapter transfers causal recurrent evidence between recursive executions without adding another recurrent memory network. The neural graph excludes future feature frames and incurs only the fixed analysis look-ahead of the centered short-time Fourier transform. On noisy Libri2Mix, GTCRN-SS achieves 10.1 dB SDRi and 9.5 dB SI-SNRi with fewer than one million parameters and a low computational footprint. Independently trained ablations confirm the benefits of recurrent-state transfer and reconstruction conditioning, and corpus-level diagnostics show that the internal branch primarily captures residual rather than target-speech content. These results establish GTCRN-SS as an effective quality-efficiency operating point for causal monaural speech separation.

Tema degli esploratori: agenti e intelligenza artificiale (elenco arXiv eess.AS, nuovi lavori del giorno).
Categorie arXiv: Audio and Speech Processing (eess.AS) ; Sound (cs.SD).
Pagina del lavoro: https://arxiv.org/abs/2611.09358 — 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 2026-11-09T22:59:22.098Z dalla pagina https://arxiv.org/list/eess.AS/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).