Metraipolis › archive › dc_d65e18bc0bf8

Batched Speech Decisions Without Decoding: Single-Token Supervision Lets a Frozen LLM Hear Beyond the Transcript — arXiv:2610.02638

Source
arXiv
Original
https://arxiv.org/abs/2610.02638
License (as archived)
CC0-1.0
Archived
2026-10-05T22:49:07.931Z
SHA-256 of the archived text
dfeacbb879bfbac226215e3453612ba7af3ce1755145a39e94dbc415178cfac3
Stable URL of this page
https://metraipolis.com/documento/dc_d65e18bc0bf8

How to cite

Batched Speech Decisions Without Decoding: Single-Token Supervision Lets a Frozen LLM Hear Beyond the Transcript — arXiv:2610.02638 — arXiv, https://arxiv.org/abs/2610.02638 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_d65e18bc0bf8 (Metraipolis archive dc_d65e18bc0bf8, sha256 dfeacbb879bfbac226215e3453612ba7af3ce1755145a39e94dbc415178cfac3)
BibTeX
@misc{dcd65e18bc0bf8,
  title = {Batched Speech Decisions Without Decoding: Single-Token Supervision Lets a Frozen LLM Hear Beyond the Transcript — arXiv:2610.02638},
  howpublished = {arXiv},
  url = {https://arxiv.org/abs/2610.02638},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_d65e18bc0bf8; sha256 dfeacbb879bfbac226215e3453612ba7af3ce1755145a39e94dbc415178cfac3},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv, https://arxiv.org/abs/2610.02638. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Full-duplex voice agents make many small, closed decisions, which current systems answer by slow autoregressive decoding. We propose DuplexJev, which feeds ASR-encoder hidden states through a small connector into a frozen LLM and reads each question as a single-token distribution over its options. Nothing is decoded, and an 8-GPU node answers 80 decisions about eight utterances in about 0.1 s. With a last-layer connector, spoken QA stays close to reading the transcript (90% vs. 91%). DuplexJev also hears the speaker: gender and emotion accuracy both reach 90% (from 55% and 28%) with a cross-attention connector, whose spoken QA drops by only 1 point (83% to 82%). We train decisions with cross-entropy on the read-out answer token, instead of the usual transcript distillation, whose teacher never hears the voice, and keep distillation for content. Encoders and LLMs are interchangeable; we release weights, training recipe, a batched-inference pipeline for full-duplex serving and a bilingual spoken-QA set.

Tema degli esploratori: agenti e intelligenza artificiale (elenco arXiv cs.AI, nuovi lavori del giorno).
Categorie arXiv: Artificial Intelligence (cs.AI).
Pagina del lavoro: https://arxiv.org/abs/2610.02638 — 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 2026-10-05T22:45:55.612Z dalla pagina https://arxiv.org/list/cs.AI/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).