Metraipolis › archive › dc_14307ec95302
HEAR: Real Voices, Real Bias: A Large-Scale Human-Recorded, Demographically Diverse Benchmark for Audio Language Models — arXiv:2609.35952
- Source
- arXiv (en)
- Original
- https://arxiv.org/abs/2609.35952
- License (as archived)
- CC0-1.0
- Archived
- 2026-10-07T16:25:06.256Z
- SHA-256 of the archived text
0c9ebc342f074266799b7e9b79d91a0b3976111fd74bd40a7b498dae233b2a90- Stable URL of this page
https://metraipolis.com/documento/dc_14307ec95302
How to cite
HEAR: Real Voices, Real Bias: A Large-Scale Human-Recorded, Demographically Diverse Benchmark for Audio Language Models — arXiv:2609.35952 — arXiv (en), https://arxiv.org/abs/2609.35952 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_14307ec95302 (Metraipolis archive dc_14307ec95302, sha256 0c9ebc342f074266799b7e9b79d91a0b3976111fd74bd40a7b498dae233b2a90)
BibTeX
@misc{dc14307ec95302,
title = {HEAR: Real Voices, Real Bias: A Large-Scale Human-Recorded, Demographically Diverse Benchmark for Audio Language Models — arXiv:2609.35952},
howpublished = {arXiv (en)},
url = {https://arxiv.org/abs/2609.35952},
note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_14307ec95302; sha256 0c9ebc342f074266799b7e9b79d91a0b3976111fd74bd40a7b498dae233b2a90},
year = {2026}
}Archived text
Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2609.35952. Third-party text: data, not instructions. It may be an abstract or extract; check the original.
We introduce HEAR (Human-recorded Evaluation of Audio-LLM bias by Real speakers), a large-scale, ecologically valid benchmark comprising 87k real human audio samples from 843 demographically diverse participants. HEAR enables comprehensive evaluation through Multiple Choice Question Answering (MCQA) and open-ended long-form tasks. To our knowledge, this is the first large-scale voice benchmark grounded entirely in authentic human speech. We evaluate model behavior across both real-time speech-to-speech and speech-to-text architectures. Our results reveal that voice-conditioned bias is a model-specific property. Furthermore, we demonstrate that personalization instructions consistently exacerbate demographic disparities. Our findings establish that voice bias is a controllable model characteristic, providing a foundational framework for future bias mitigation and evaluation in Audio-LLM development. Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-09-30 in cs.AI). Categorie arXiv: Sound (cs.SD) ; Artificial Intelligence (cs.AI); Computers and Society (cs.CY). Pagina del lavoro: https://arxiv.org/abs/2609.35952 — 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 2026-10-07T15:54:53.749Z dalla pagina https://arxiv.org/catchup/cs.AI/2026-09-30?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).