AI models tested for Arabic

Whisper-base fine-tuned for Quranic Arabic recitation, reporting 5.75% WER on its evaluation set. Narrow by design — it targets recitation rather than conversational Arabic — but it is the most-adopted open Arabic speech model outside the general Whisper checkpoints, and underpins Tarteel's memorisation app. A tiny variant is also published.

2.35B Arabic-first speech recognition trained on 300k+ hours, naming Gulf, Egyptian, Levantine and Maghrebi coverage plus Arabic-English code-switching. Its card claims the top place on the Open Universal Arabic ASR Leaderboard — a leaderboard run by Elm, a Saudi company. Custom AudarAI Community Licence, not open source.

Arabic-specialised speech-to-text model (Jul 2026) built on Cohere's general transcribe checkpoint. One of the most-downloaded dedicated Arabic ASR repositories on Hugging Face, and unusually for an Arabic ASR release it carries a permissive licence. The repo is gated behind a contact-sharing agreement.

Arabic ASR trained with weak supervision on 30K+ hours; the paper claims best-in-class accuracy across 18 dialects. No weights are published — there is no CNTXT organisation or Munsit repository on Hugging Face — and the model is commercial API-only, so the dialect claims cannot be independently checked.

State-of-the-art open speech recognition covering 99+ languages and the default open ASR baseline for Arabic products, though Arabic is one language among many rather than a focus and dialectal Arabic remains its weak point. Note the licence inconsistency: this model card states Apache 2.0 while the openai/whisper GitHub repo and the large-v3-turbo card state MIT.