Metraipolis › archive › dc_875d0bc040be

Almieyar: A Culturally Grounded Benchmark for Multi-Dialect Arabic Speech Recognition — arXiv:2609.35564

Source: arXiv (en)
Original: https://arxiv.org/abs/2609.35564
License (as archived): CC0-1.0
Archived: 2026-10-07T21:03:46.536Z
SHA-256 of the archived text: 1175372a9245a51c5204c5816b5727f106061c4547c28e5083549e8a28b60418
Stable URL of this page: https://metraipolis.com/documento/dc_875d0bc040be

How to cite

Almieyar: A Culturally Grounded Benchmark for Multi-Dialect Arabic Speech Recognition — arXiv:2609.35564 — arXiv (en), https://arxiv.org/abs/2609.35564 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_875d0bc040be (Metraipolis archive dc_875d0bc040be, sha256 1175372a9245a51c5204c5816b5727f106061c4547c28e5083549e8a28b60418)

BibTeX

@misc{dc875d0bc040be,
  title = {Almieyar: A Culturally Grounded Benchmark for Multi-Dialect Arabic Speech Recognition — arXiv:2609.35564},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2609.35564},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_875d0bc040be; sha256 1175372a9245a51c5204c5816b5727f106061c4547c28e5083549e8a28b60418},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2609.35564. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Arabic speech technology has largely focused on Modern Standard Arabic, leaving the living dialects spoken by hundreds of millions under-served. We introduce ALMIEYAR, a culturally grounded ASR benchmark covering 17 Arabic dialects across six families, built entirely from newly recorded speech unseen by existing models. Dialect-community coordinators selected culturally relevant images across 10 topics, and native speakers described them through five structured scenarios, yielding approximately 50 minutes per dialect (13.7 hours total). We benchmark 12 state-of-the-art ASR systems zero-shot, including GPT-4o-transcribe, Voxtral-Mini-4B, Fanar-STT-LF, Whisper, SeamlessM4T-v2, and wav2vec2-based models. GPT-4o-transcribe achieves the lowest overall WER at 35.0%, followed by Voxtral-Mini-4B, Fanar-STT-LF, and Whisper-Large-v3 at 41.1%, 45.9%, and 49.5%, respectively, indicating substantial remaining errors across Arabic dialect communities. Performance varies considerably across dialect groups, with no model performing uniformly best across all groups. WER alone also obscures dialectal ASR behaviour: wav2vec2-based models show large WER/CER gaps, where character-level agreement remains much higher than word-level accuracy, motivating joint WER/CER reporting. ALMIEYAR provides a unified benchmark for culturally grounded Arabic ASR evaluation, including the first published benchmark for Ahwazi Arabic.

Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-09-29 in cs.AI).
Categorie arXiv: Computation and Language (cs.CL) ; Artificial Intelligence (cs.AI).
Pagina del lavoro: https://arxiv.org/abs/2609.35564 — 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 2026-10-07T20:29:03.397Z dalla pagina https://arxiv.org/catchup/cs.AI/2026-09-29?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).