Metraipolis › archive › dc_ebe2eef82c44

Separate First, Then Associate: A Two-Stage Approach for Real-World Audio-Visual Speech Enhancement — arXiv:2608.14812

Source: arXiv (en)
Original: https://arxiv.org/abs/2608.14812
License (as archived): CC0-1.0
Archived: 2026-10-19T02:19:07.466Z
SHA-256 of the archived text: 021d727dfc95d2a3692bcb933ec706bd19f34c8a8e4200abe9de2490a7c2a0f5
Stable URL of this page: https://metraipolis.com/documento/dc_ebe2eef82c44

How to cite

Separate First, Then Associate: A Two-Stage Approach for Real-World Audio-Visual Speech Enhancement — arXiv:2608.14812 — arXiv (en), https://arxiv.org/abs/2608.14812 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_ebe2eef82c44 (Metraipolis archive dc_ebe2eef82c44, sha256 021d727dfc95d2a3692bcb933ec706bd19f34c8a8e4200abe9de2490a7c2a0f5)

BibTeX

@misc{dcebe2eef82c44,
  title = {Separate First, Then Associate: A Two-Stage Approach for Real-World Audio-Visual Speech Enhancement — arXiv:2608.14812},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2608.14812},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_ebe2eef82c44; sha256 021d727dfc95d2a3692bcb933ec706bd19f34c8a8e4200abe9de2490a7c2a0f5},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2608.14812. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Audio-visual speech enhancement (AVSE) aims at extracting target speech from multi-speaker mixtures by exploiting visual cues. Although recent studies have reported strong performance on simulated datasets, the performance, however, often drops dramatically when they are applied to real-world audio-visual recordings. To bridge this gap, the Real-World AVSE Challenge held in the ISCSLP 2026 conference calls for participants to design a practical solution for AVSE under real-world conditions, where speaker overlap, acoustic interferences, room reverberation and visual degradations naturally co-exist. In our submission to the challenge, we propose a decoupled separation-then-association approach. It consists of two stages: a separation stage in which a trained, audio-only model (i.e., not using visual cues) is used to separate input multi-speaker mixture to individual speaker signals, followed by an association stage, where an audio-visual CLIP model is used to identify the separated speech signal with the highest similarity with the target speaker's facial video via cross-modal similarity matching. Evaluation results on the challenge dataset show the effectiveness of our proposed approach.

Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-08-18 in eess.AS).
Categorie arXiv: Audio and Speech Processing (eess.AS).
Pagina del lavoro: https://arxiv.org/abs/2608.14812 — 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 2026-10-19T01:56:37.390Z dalla pagina https://arxiv.org/catchup/eess.AS/2026-08-18?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).