Oossa

Falcon ASR speech model released for Arabic and four other languages

Hugging Face unveils a 1.6 B‑parameter model that lowers Arabic word error rate to 20.9% and supports English, French, Spanish and Portuguese.

NoteBy Published by Oossa: 1 min read

Hugging Face has launched Falcon‑ASR, a 1.6 billion‑parameter speech‑recognition model built by the Technology Innovation Institute in Abu Dhabi. It targets Arabic, especially the Emirati dialect, and also transcribes English, French, Spanish and Portuguese. In the Open Universal Arabic ASR Leaderboard the model achieved an average word error rate (WER) of 20.92%, beating the previous best published score of 23.17%.

Why it matters

Arabic speakers, particularly in the UAE, can now get more accurate automated transcriptions for meetings and calls.

Was this article useful?
Share

Read next

Oossa · Newsletter

The week in AI, explained

Every Monday: the stories worth knowing, in plain language. Free, no spam.