Metraipolis › archive › dc_d2e73111f9c2

INSPIRE: A Benchmark for Instruction-Aware Speech Retrieval — arXiv:2608.16203

Source
arXiv (en)
Original
https://arxiv.org/abs/2608.16203
License (as archived)
CC0-1.0
Archived
2026-10-18T19:01:28.603Z
SHA-256 of the archived text
f3eea05e4909b633cc6dc81e489cc0c615c944d2a8bbeb05e826f0bf1e40d6a0
Stable URL of this page
https://metraipolis.com/documento/dc_d2e73111f9c2

How to cite

INSPIRE: A Benchmark for Instruction-Aware Speech Retrieval — arXiv:2608.16203 — arXiv (en), https://arxiv.org/abs/2608.16203 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_d2e73111f9c2 (Metraipolis archive dc_d2e73111f9c2, sha256 f3eea05e4909b633cc6dc81e489cc0c615c944d2a8bbeb05e826f0bf1e40d6a0)
BibTeX
@misc{dcd2e73111f9c2,
  title = {INSPIRE: A Benchmark for Instruction-Aware Speech Retrieval — arXiv:2608.16203},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2608.16203},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_d2e73111f9c2; sha256 f3eea05e4909b633cc6dc81e489cc0c615c944d2a8bbeb05e826f0bf1e40d6a0},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2608.16203. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Existing speech retrieval systems rely on fixed similarity matching and cannot adapt to diverse user intents. We introduce INSPIRE, the first benchmark for instruction-aware speech retrieval, in which natural-language instructions dynamically specify relevance criteria, including semantic content, speaker identity, speaking style, environmental sounds, and their combinations. We evaluate four retrieval paradigms: large audio-language models, cascaded pipelines, self-supervised speech models, and contrastive audio-language models. Our results reveal that no current method robustly handles all retrieval intents. Text-based approaches perform relatively better at semantic retrieval but struggle with paralinguistic attributes, while speech-based models are moderately better at capturing acoustic properties but falter at following instructions. These findings highlight the need for unified architectures capable of instruction-aware speech retrieval.

Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-08-18 in cs.CL).
Categorie arXiv: Sound (cs.SD) ; Computation and Language (cs.CL); Audio and Speech Processing (eess.AS).
Pagina del lavoro: https://arxiv.org/abs/2608.16203 — 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 2026-10-18T18:06:57.930Z dalla pagina https://arxiv.org/catchup/cs.CL/2026-08-18?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).