Metraipolis › archive › dc_fc5cdbc17d26

IndicFDB: Benchmarking Full-Duplex Voice Agents across Indian Languages — arXiv:2609.31967

Source
arXiv (en)
Original
https://arxiv.org/abs/2609.31967
License (as archived)
CC0-1.0
Archived
2026-10-07T21:49:56.122Z
SHA-256 of the archived text
1caa6e7cc1deda6f708d1cfa74d54b77efb36c9e9281ebbf88181f6f63c13d84
Stable URL of this page
https://metraipolis.com/documento/dc_fc5cdbc17d26

How to cite

IndicFDB: Benchmarking Full-Duplex Voice Agents across Indian Languages — arXiv:2609.31967 — arXiv (en), https://arxiv.org/abs/2609.31967 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_fc5cdbc17d26 (Metraipolis archive dc_fc5cdbc17d26, sha256 1caa6e7cc1deda6f708d1cfa74d54b77efb36c9e9281ebbf88181f6f63c13d84)
BibTeX
@misc{dcfc5cdbc17d26,
  title = {IndicFDB: Benchmarking Full-Duplex Voice Agents across Indian Languages — arXiv:2609.31967},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2609.31967},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_fc5cdbc17d26; sha256 1caa6e7cc1deda6f708d1cfa74d54b77efb36c9e9281ebbf88181f6f63c13d84},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2609.31967. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Full-duplex voice agents must handle pauses, take turns, backchannel, and respond to user interruptions in real time. Full-Duplex-Bench evaluates these behaviors, but its English-only corpus and reliance on word-timestamped ASR and an English-prompted LLM judge make it difficult to extend to Indian languages. We introduce IndicFDB, which extends it to ten languages spoken in India with 12,350 samples, nearly 17 times as many as the original. We address three challenges: finding conversational events in multilingual speech, evaluating their timing without reliable word-level alignment, and judging responses across languages. We mine pause handling, turn taking, and backchanneling samples from roughly 50,000 hours of channel-separated conversations using voice activity detection (VAD), and construct human-validated synthetic user interruption samples. Language-independent VAD heuristics evaluate timing, while an open-weight transcription and translation pipeline converts responses to English for LLM ratings of relevance and quality. Across seven voice agents, commercial APIs show unexpectedly consistent behavior across languages but are either fast or robust to pauses, never both, while monolingual open full-duplex models expose further tradeoffs among backchanneling, response quality, and latency.

Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-09-29 in cs.CL).
Categorie arXiv: Computation and Language (cs.CL).
Pagina del lavoro: https://arxiv.org/abs/2609.31967 — 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 2026-10-07T21:27:08.377Z dalla pagina https://arxiv.org/catchup/cs.CL/2026-09-29?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).