Metraipolis › archive › dc_9f2e071dfa40

VoiceTrace: A Benchmark and Retrieval Framework for Who-Said-What Speech Retrieval — arXiv:2609.18521

Source
arXiv (en)
Original
https://arxiv.org/abs/2609.18521
License (as archived)
CC0-1.0
Archived
2026-10-15T18:28:09.077Z
SHA-256 of the archived text
8982d5d38519786cd9b69435b6eddadbc134ba5a677082151db2498bbee4b0c2
Stable URL of this page
https://metraipolis.com/documento/dc_9f2e071dfa40

How to cite

VoiceTrace: A Benchmark and Retrieval Framework for Who-Said-What Speech Retrieval — arXiv:2609.18521 — arXiv (en), https://arxiv.org/abs/2609.18521 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_9f2e071dfa40 (Metraipolis archive dc_9f2e071dfa40, sha256 8982d5d38519786cd9b69435b6eddadbc134ba5a677082151db2498bbee4b0c2)
BibTeX
@misc{dc9f2e071dfa40,
  title = {VoiceTrace: A Benchmark and Retrieval Framework for Who-Said-What Speech Retrieval — arXiv:2609.18521},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2609.18521},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_9f2e071dfa40; sha256 8982d5d38519786cd9b69435b6eddadbc134ba5a677082151db2498bbee4b0c2},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2609.18521. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Speech retrieval has become increasingly important as spoken content continues to grow across meetings, lectures, podcasts, and videos. Existing benchmarks and models have advanced semantic search over spoken content, but largely focus on \emph{what} is said while overlooking \emph{who} says it. In many real-world scenarios, however, users need to retrieve speech based jointly on semantic content and a target speaker, where the speaker may be specified naturally through a reference speech utterance rather than a predefined identity. To address this gap, we introduce \textbf{VoiceTrace-Bench}, a benchmark for hybrid speech retrieval in which each query combines text specifying \emph{what} to retrieve with reference speech specifying \emph{who} to retrieve. This setting requires models to integrate complementary semantic and speaker information directly from heterogeneous query inputs. Motivated by the joint audio-text modeling capabilities of audio-language models (ALMs), we develop \textbf{VoiceTrace}, a two-stage retrieval framework consisting of \textbf{VoiceTrace-Emb}, an embedding model that learns unified representations for efficient large-scale retrieval, and \textbf{VoiceTrace-Reranker}, a reranking model that jointly examines each query--candidate pair for fine-grained relevance estimation. Experiments show that VoiceTrace achieves state-of-the-art performance on established semantic speech retrieval benchmarks, while substantially outperforming cascade-based approaches on VoiceTrace-Bench, demonstrating its effectiveness for both conventional semantic retrieval and the new hybrid retrieval setting.

Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-09-17 in eess.AS).
Categorie arXiv: Sound (cs.SD) ; Artificial Intelligence (cs.AI); Audio and Speech Processing (eess.AS).
Pagina del lavoro: https://arxiv.org/abs/2609.18521 — 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 2026-10-15T17:57:43.709Z dalla pagina https://arxiv.org/catchup/eess.AS/2026-09-17?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).