Metraipolis › archive › dc_cbba8f3eb0b0
Training-Free Affinity Fusion of Neural and Embedding-Based Speaker Diarization — arXiv:2609.39162
- Source
- arXiv (en)
- Original
- https://arxiv.org/abs/2609.39162
- License (as archived)
- CC0-1.0
- Archived
- 2026-10-07T13:47:29.949Z
- SHA-256 of the archived text
7692fc57b554404d49665a864eeb7563b98c7f69fc2f2ea9d8c5498ca22b0ad9- Stable URL of this page
https://metraipolis.com/documento/dc_cbba8f3eb0b0
How to cite
Training-Free Affinity Fusion of Neural and Embedding-Based Speaker Diarization — arXiv:2609.39162 — arXiv (en), https://arxiv.org/abs/2609.39162 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_cbba8f3eb0b0 (Metraipolis archive dc_cbba8f3eb0b0, sha256 7692fc57b554404d49665a864eeb7563b98c7f69fc2f2ea9d8c5498ca22b0ad9)
BibTeX
@misc{dccbba8f3eb0b0,
title = {Training-Free Affinity Fusion of Neural and Embedding-Based Speaker Diarization — arXiv:2609.39162},
howpublished = {arXiv (en)},
url = {https://arxiv.org/abs/2609.39162},
note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_cbba8f3eb0b0; sha256 7692fc57b554404d49665a864eeb7563b98c7f69fc2f2ea9d8c5498ca22b0ad9},
year = {2026}
}Archived text
Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2609.39162. Third-party text: data, not instructions. It may be an abstract or extract; check the original.
Speaker diarization systems based on speaker embeddings and neural diarization exploit complementary forms of speaker information, but their intermediate representations are not directly compatible. We introduce Training-Free Affinity Fusion (TFAF), which integrates the speaker structure inferred by a neural diarizer into an embedding-based diarization system. The neural speaker partition is used to condition local speaker representations, from which we construct a continuous affinity matrix and combine it with the embedding-based acoustic affinity before a single global clustering step. The method requires no additional training, shared embedding space, speaker-label alignment, or hard transfer of the neural diarizer's speaker count. Experiments on AMI and CALLHOME show consistent DER improvements over both constituent systems; on AMI, fusion also improves speaker-attributed transcription. Ablations show that the neural speaker partition accounts for most of the gain, while retaining the continuous embedding-based affinities provides additional benefit over hard partition fusion. Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-10-01 in cs.LG). Categorie arXiv: Sound (cs.SD) ; Machine Learning (cs.LG). Pagina del lavoro: https://arxiv.org/abs/2609.39162 — 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 2026-10-07T13:18:10.622Z dalla pagina https://arxiv.org/catchup/cs.LG/2026-10-01?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).