Open Universal Arabic Asr Leaderboard

Open Universal Arabic Asr Leaderboard

Compares open-source speech recognition models on Arabic across several dialects, in a Gradio page. For researchers and developers choosing an Arabic speech-to-text model.

Made by msalhab96 on Hugging Face. These notes are Mindket's, written from its Space page and README; the app's own files and license are on Hugging Face.

About

What it is: This Space is a benchmark page for open-source speech recognition models that handle Arabic in several dialects. The record names six models and one dataset, UBC-NLP/Casablanca. The source gives only a one-line description, so the page itself holds the details.

Who it suits: It suits researchers and developers who need to compare Arabic speech-to-text models before picking one.

Industry | Who uses it | How they use it | Example
Speech research | Researchers | Compare open models on Arabic dialects | A researcher reads how the listed models differ
Software | Developers | Choose an Arabic speech model for an app | A developer checks the options before testing one
Media | Transcription teams | See which open models are in the comparison | A team looks for a model that fits its audio
Linguistics | Language researchers | Look at how recognition models treat dialect differences | A linguist reads the benchmark for dialect coverage

You can skip it if you need speech recognition for a language other than Arabic.

Where to use it: You use it in these places.

• A web browser, on the Space's page on Hugging Face
• A local copy on Windows, macOS or Linux, which needs Python

Use case | Who | What they get back
Compare models | Researchers | The benchmark view of the open Arabic models it lists
Shortlist a model | Developers | A starting set of models to test on their own audio
Study the setup | Developers | The app.py file and the dataset the record names

How to set it up: Online use needs only a browser. A local copy needs Python and the packages the repository lists.

• A web browser, for the online page
• Python with pip, for a local copy
• Python packages from requirements.txt

1. Open the Space's page on Hugging Face to use it online, with nothing to install.
2. For a local copy, install the Hugging Face Hub package.
3. Download the Space's files into a folder and move into it.
4. Install Gradio 5.49.1 and the packages in requirements.txt.
5. Start the app.

pip install -U huggingface_hub
hf download msalhab96/open_universal_arabic_asr_leaderboard_all --repo-type space --local-dir open_universal_arabic_asr_leaderboard_all
cd open_universal_arabic_asr_leaderboard_all
pip install gradio==5.49.1
pip install -r requirements.txt
python app.py

How to use it: Open the Space's page and read the benchmark. For a local copy, start the app with the command below and open the page it serves in your browser.

python app.py

Example: You open the page to see which open Arabic speech models are compared. You read the benchmark for the models that interest you and note them for a test on your own audio.

For newcomers and experts: ASR stands for automatic speech recognition, which turns spoken audio into text. Arabic has many dialects, and a model can do well on one and badly on another. Start by opening the Space's page and reading the benchmark.

Under the hood, the app is built with Gradio 5.49.1 and runs on cpu-basic hardware at Hugging Face. The record names asafaya's hubert-large Arabic transcribe model, two Audar-ASR models from audarai, two Cohere transcribe models and Meta's mms-1b-all. The source does not say whether the page runs these models live.

You open the Space's page
You read the Arabic benchmark
You compare the listed models
You test a shortlist on your audio

Good to know: The source is brief, so check the page itself for the details of the benchmark.

• License: apache-2.
• Cost to run: none named. The Space uses cpu-basic hardware at Hugging Face.
• Data it reaches: the record names the dataset UBC-NLP/Casablanca and six models, which come from Hugging Face.
• Not for: languages other than Arabic, since the description covers Arabic only.

On Hugging Face, 13 November 2026

• Likes: 16
• Built with: gradio
• License: apache-2.0

Links

• Space: https://huggingface.co/spaces/msalhab96/open_universal_arabic_asr_leaderboard_all
• Files: https://huggingface.co/spaces/msalhab96/open_universal_arabic_asr_leaderboard_all/tree/main

Try it

Open the Space in your browser, or copy it to your own Hugging Face account from its page. Nothing in this kit runs by itself.