🔍 Read the full analysis: A First Look At Falcon ASR And Its Speech Recognition Approach on ThorstenMeyerAI.com
Get business pricing on tech for your team
- Business-only prices and quantity discounts
- Tax-exempt purchasing
- Multiple users, one account, clear invoices
TL;DR
Abu Dhabi’s Technology Innovation Institute has introduced Falcon-ASR, a 1.6-billion-parameter speech recognition model focused on Arabic, including Emirati speech. TII reports a 20.92% average word error rate across six Arabic test sets and a 22.73% WER in its internal Emirati evaluation; the results are developer-reported and the model’s real-world performance remains to be established.
The Technology Innovation Institute (TII) in Abu Dhabi has introduced Falcon-ASR, a 1.6-billion-parameter speech recognition model designed for Arabic, with a particular focus on the Emirati dialect, as described in the original analysis. TII reports a 20.92% average word error rate across six Arabic test sets and says the model had the lowest error rates among systems in its internal Emirati evaluation; those results have not been independently verified in the material provided.
On the six Arabic test sets used by the Open Universal Arabic ASR Leaderboard, TII reports an equal-weight average word error rate (WER) of 20.92%. The institute compared that with the best published average in a leaderboard snapshot it checked on September 30, 2026: 23.17%. That is a 2.25 percentage-point difference. WER measures word-level transcription errors, with lower scores indicating fewer errors. TII says it followed the leaderboard protocol and used its pinned manifests.
For Emirati speech, TII reports 22.73% WER and 10.19% character error rate (CER) in an internal evaluation using held-out Emirati and Gulf recordings with human-validated transcripts. The institute says these were the lowest scores among the systems it compared, and that the next-best WER, from Qwen3-Omni, was 4.07 percentage points higher. The announcement does not provide the full comparison or detailed results for each recording group.
TII says Falcon-ASR can transcribe Arabic, English, French, Spanish and Portuguese using the same model weights, without a required language flag. It was trained on Emirati, Modern Standard Arabic, other Gulf and Arabic dialects, and English, according to the institute. TII also reports word-level timestamps, which associate transcribed words with their positions in an audio recording. The model is available through a Hugging Face demo; API access and native applications are planned, with no release dates given.
Why Emirati Speech Results Matter
Speech recognition systems can be harder to assess on everyday dialects than on more formal speech. Arabic varies across regions and situations, while dialect-specific transcribed material is less available than material for Modern Standard Arabic. TII’s reported Emirati evaluation addresses a practical question for users building tools for calls, meetings and informal recordings: whether a model can handle speech beyond formal or broadcast-style Arabic.
The figures are relevant as an initial benchmark, not a guarantee for every speaker or recording. If performance holds up across more dialects and real-world conditions, Falcon-ASR could support transcription workflows where speakers use Gulf varieties or switch between languages. Word-level timestamps may also help users locate passages in longer recordings. The announced demo gives people a way to try specific audio, but its results cannot by themselves establish broad accuracy.
Arabic speech recognition software
As an affiliate, we earn on qualifying purchases.
As an affiliate, we earn on qualifying purchases.
How TII Measured Falcon-ASR
The Arabic leaderboard comparison uses six test sets and gives each set equal weight in its average. TII says the competitor figures came from published leaderboard results and that it checked a September 30, 2026 snapshot. Its reported 20.92% score is therefore a comparison against that dated set of results, rather than a live ranking or a comparison with every available speech recognition system. The source material does not list Falcon-ASR’s individual scores on the six sets.
TII describes the Emirati result as an internal evaluation and also points to the UAE subset of the public Casablanca dataset. The institute says Falcon-ASR builds on its earlier Falcon3-Audio work. Separately, it reports a mean WER of 5.74% on seven public English test sets used by the Hugging Face Open ASR Leaderboard. These are results reported by the developer; the supplied information does not describe an independent replication.
““Our aim is to transcribe the words people use in everyday speech, including dialectal forms and switches between languages.””
— Technology Innovation Institute
As an affiliate, we earn on qualifying purchases.
What the Evaluation Leaves Open
The announcement does not give a complete breakdown of results by Arabic test set, dialect, speaker or recording condition. It also does not state the size or full composition of the internal Emirati evaluation or list every system included in that comparison. Those details would help readers judge how widely the reported WER and CER scores apply.
The leaderboard result is tied to a snapshot checked on September 30, 2026, and later published scores may change the comparison. The reported figures are from TII, and no independent replication is described in the source material. It remains unclear how Falcon-ASR will perform on speech that differs from the evaluation recordings, including across accents, noisy environments and everyday applications.
Emirati dialect transcription tool
As an affiliate, we earn on qualifying purchases.
As an affiliate, we earn on qualifying purchases.
Demo Access and Planned Releases
Users can try Falcon-ASR through TII’s Hugging Face Demo Space, which the institute says accepts users’ own recordings for transcription. TII has also said API access and native applications are planned, but has not announced release dates. Their availability could make the model easier to incorporate into products and existing workflows.
For a clearer picture of performance, useful next disclosures would include per-test-set and per-dialect scores, more information about the internal evaluation, and independent testing. Until then, the demo lets users inspect outputs on particular recordings, while the published benchmark results describe performance only on the evaluation material identified by TII.
language model for Arabic transcription
As an affiliate, we earn on qualifying purchases.
As an affiliate, we earn on qualifying purchases.
Key Questions
What is Falcon-ASR?
Falcon-ASR is a 1.6-billion-parameter speech recognition model introduced by Abu Dhabi’s Technology Innovation Institute. TII says it supports Arabic, English, French, Spanish and Portuguese.
What Arabic benchmark result did TII report?
TII reports a 20.92% average WER across six Arabic test sets used by the Open Universal Arabic ASR Leaderboard. It compared that with a 23.17% best published average in the snapshot checked on September 30, 2026.
How did Falcon-ASR perform on Emirati speech?
TII reports 22.73% WER and 10.19% CER on an internal evaluation of held-out Emirati and Gulf recordings with human-validated transcripts. The institute says these were the lowest scores among the systems it compared.
Can the reported results be treated as independent verification?
No independent replication is described in the source material. The benchmark and internal evaluation figures are reported by TII, and the announcement does not provide full per-set results or all details of the Emirati comparison.
How can people try Falcon-ASR?
TII says the model can be tried through its Hugging Face Demo Space. API access and native applications are planned, but no release dates have been announced.
Primary source: Hugging Face · via ThorstenMeyerAI.com
Fall Picks
fall essentials
As an affiliate, we earn on qualifying purchases.
