Metraipolis › archive › dc_f8bb9993205c

FD-VAD: Semantic Endpoint Detection for Streaming Full-Duplex Speech — arXiv:2609.35791

Source
arXiv (en)
Original
https://arxiv.org/abs/2609.35791
License (as archived)
CC0-1.0
Archived
2026-10-07T16:56:15.499Z
SHA-256 of the archived text
4805a573bac335317ca39ae34a6056b1ebfee50b99b85d76531db2d656712ed3
Stable URL of this page
https://metraipolis.com/documento/dc_f8bb9993205c

How to cite

FD-VAD: Semantic Endpoint Detection for Streaming Full-Duplex Speech — arXiv:2609.35791 — arXiv (en), https://arxiv.org/abs/2609.35791 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_f8bb9993205c (Metraipolis archive dc_f8bb9993205c, sha256 4805a573bac335317ca39ae34a6056b1ebfee50b99b85d76531db2d656712ed3)
BibTeX
@misc{dcf8bb9993205c,
  title = {FD-VAD: Semantic Endpoint Detection for Streaming Full-Duplex Speech — arXiv:2609.35791},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2609.35791},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_f8bb9993205c; sha256 4805a573bac335317ca39ae34a6056b1ebfee50b99b85d76531db2d656712ed3},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2609.35791. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Natural turn-taking in full-duplex voice interaction requires determining from partial speech whether a pause reflects hesitation or a completed conversational intent. Acoustic voice activity detection lacks this semantic information, while cascaded ASR-based endpointing introduces transcription dependence and additional processing stages. We formulate semantic endpoint detection as a causal audio-language reasoning task and introduce FD-VAD, an ASR-free streaming endpointer that maps bounded causal audio windows directly to Continue/Stop decisions. FD-VAD combines a frozen speech encoder with a lightweight modality adapter and a parameter-efficiently adapted language model, using a last-chunk training objective for streaming inference. We further introduce confidence-gated endpoint commitment to control interruption versus delay and boundary-focused hard-negative sampling to improve decisions around ambiguous turn boundaries. Across in-domain and conversational evaluations, FD-VAD outperforms strong streaming and non-streaming semantic turn classifiers, and achieves the highest EOT recall among qualifying systems on TurnBench dev set $0.853$ (at FP<=0.10) in a zero-shot setting. These results show that semantic endpointing can be performed directly from streaming audio without intermediate ASR or dialogue state tracking.

Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-09-30 in cs.CL).
Categorie arXiv: Computation and Language (cs.CL) ; Audio and Speech Processing (eess.AS).
Pagina del lavoro: https://arxiv.org/abs/2609.35791 — 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 2026-10-07T16:27:03.143Z dalla pagina https://arxiv.org/catchup/cs.CL/2026-09-30?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).