TII Releases Falcon-ASR: 1.6B Arabic Speech Model Achieves 20.92% WER
The Technology Innovation Institute (TII) in Abu Dhabi has unveiled Falcon-ASR, a 1.6B-parameter Arabic speech recognition model. Designed with a focus on the Emirati dialect, it also transcribes English, French, Spanish, and Portuguese. In TII’s evaluation, Falcon-ASR achieves a 20.92% average word error rate (WER) across six Arabic test sets—outperforming the best published entry in TII’s leaderboard snapshot by 2.25 percentage points. However, for developers building Arabic voice products, a key limitation remains: the model is currently available only as a hosted demo.
TL;DR
- Size: 1.6B parameters. Maximum audio context is not disclosed (the demo accepts clips up to 60 seconds and 20 MB).
- Runs on: Hardware requirements are not disclosed. No weights or quantized variants have been published. Access is limited to a Hugging Face demo, with API and native apps planned.
- Performance: Achieves the lowest Arabic and Emirati WER in TII’s own comparisons, alongside solid English results.
- Best result: 20.92% average WER across six Arabic test sets.
via MarkTechPost