Mindket

Open Universal Arabic Asr Leaderboard by elmresearchcenter

Newversion 1 · updated

Overview

Compares open-source speech recognition models for Arabic dialects on one benchmark, for people who need to choose or study an Arabic ASR model.

Open Universal Arabic Asr Leaderboard by elmresearchcenter is a kit in the Research category, listed by Mindket. It is free. This is version 1, last updated on . Download it on this page, a zip of its files. Nothing in it runs by itself.

At a glance

You need
  • Python and pip on Windows, macOS or Linux
  • The Python packages in requirements.txt
  • Gradio 5.49.0, the version the Space uses
Runs in
  • A browser, by visiting the Space on Hugging Face
  • Windows, macOS or Linux with Python, for a copy on your own computer
  • A local Gradio page, once app.py is running
Start with
pip install -U huggingface_hub
hf download elmresearchcenter/open_universal_arabic_asr_leaderboard --repo-type space --local-dir open_universal_arabic_asr_leaderboard
cd open_universal_arabic_asr_leaderboard

Platform

WindowsOperating system
macOSOperating system
LinuxOperating system

About

What it is

This Space is a leaderboard for speech-to-text models that handle Arabic in several dialects. Its author presents it as a test bed for freely available ASR models that cover several Arabic dialects. It is a Gradio app that runs on basic CPU hardware at Hugging Face.

Who it suits

People who need to pick or study an Arabic speech recognition model and want models from different teams listed side by side.

IndustryWho uses itHow they use itExample
ResearchSpeech and language researchersCompare open models on Arabic dialectsLook up how a model such as Audar-ASR-V1-Turbo is placed
MediaTeams that transcribe Arabic audioShortlist a model before a pilotCheck the listed models before choosing one
SoftwareDevelopers adding speech inputPick an open model to build onRead the model list, then test the one that fits
EducationStudents of speech technologyLearn how ASR models are benchmarkedStudy the app and the models it names

People who only need a finished transcription service can skip it, because it ranks models and does not transcribe your files.

Where to use it

  • A browser, by visiting the Space on Hugging Face
  • Windows, macOS or Linux with Python, for a copy on your own computer
  • A local Gradio page, once app.py is running
Use caseWhoWhat they get back
Compare Arabic ASR modelsResearchersA leaderboard of the open models the Space lists
Shortlist a modelDevelopersNames of candidates to test further

How to set it up

  • Python and pip on Windows, macOS or Linux
  • The Python packages in requirements.txt
  • Gradio 5.49.0, the version the Space uses
  1. Install the Hugging Face hub package.
  2. Download the Space with the hf command.
  3. Move into the new folder.
  4. Install gradio at the version the Space uses, then the listed packages.
  5. Start the app with Python.
pip install -U huggingface_hub
hf download elmresearchcenter/open_universal_arabic_asr_leaderboard --repo-type space --local-dir open_universal_arabic_asr_leaderboard
cd open_universal_arabic_asr_leaderboard
pip install gradio==5.49.0
pip install -r requirements.txt
python app.py

How to use it

Online, visit the Space on Hugging Face and read the leaderboard that loads. On your own computer, run the start command and open the local page Gradio serves.

python app.py
ToolWhat it does
app.pyStarts the Gradio leaderboard
requirements.txtLists the Python packages to install
banner.pngThe banner picture used by the page

Example

You want an Arabic transcription model for audio in several dialects. You visit the Space, scan the leaderboard of the models it names, and write down the ones worth testing yourself.

For newcomers and experts

If you are new, visit the Space and look at the list of models. ASR means automatic speech recognition, which turns spoken words into text, and the leaderboard helps you see which open models are in the running for Arabic.

If you know the field, the Space names six models, including Sygma-ASR-Arabic, hubert-large-arabic-transcribe, two Audar-ASR-V1 variants and two Cohere transcribe models, plus the Casablanca dataset from UBC-NLP. The trade-off is that a leaderboard reflects one benchmark, so check it against your own audio.

  1. You visit the Gradio page
  2. The app shows the benchmark results
  3. You compare the listed models
  4. You test a shortlist on your audio

Good to know

  • Scope: it covers open-source Arabic speech recognition models across dialects
  • Hardware: the Space is set to cpu-basic, so no GPU is named
  • License: apache-2
  • Cost to run: none named
  • Data it reaches: the models and the Casablanca dataset it names on Hugging Face
  • Not for: transcribing your own files, since it is a benchmark and not a transcription tool

README and files

README

Open Universal Arabic Asr Leaderboard by elmresearchcenter

Compares open-source speech recognition models for Arabic dialects on one benchmark, for people who need to choose or study an Arabic ASR model.

Made by elmresearchcenter 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 leaderboard for speech-to-text models that handle Arabic in several dialects. Its author presents it as a test bed for freely available ASR models that cover several Arabic dialects. It is a Gradio app that runs on basic CPU hardware at Hugging Face.

Who it suits: People who need to pick or study an Arabic speech recognition model and want models from different teams listed side by side.

Industry | Who uses it | How they use it | Example
Research | Speech and language researchers | Compare open models on Arabic dialects | Look up how a model such as Audar-ASR-V1-Turbo is placed
Media | Teams that transcribe Arabic audio | Shortlist a model before a pilot | Check the listed models before choosing one
Software | Developers adding speech input | Pick an open model to build on | Read the model list, then test the one that fits
Education | Students of speech technology | Learn how ASR models are benchmarked | Study the app and the models it names

People who only need a finished transcription service can skip it, because it ranks models and does not transcribe your files.

Where to use it:

  • A browser, by visiting the Space on Hugging Face
  • Windows, macOS or Linux with Python, for a copy on your own computer
  • A local Gradio page, once app.py is running
Use case | Who | What they get back
Compare Arabic ASR models | Researchers | A leaderboard of the open models the Space lists
Shortlist a model | Developers | Names of candidates to test further

How to set it up:

  • Python and pip on Windows, macOS or Linux
  • The Python packages in requirements.txt
  • Gradio 5.49.0, the version the Space uses
  1. Install the Hugging Face hub package.
  2. Download the Space with the hf command.
  3. Move into the new folder.
  4. Install gradio at the version the Space uses, then the listed packages.
  5. Start the app with Python.
pip install -U huggingface_hub
hf download elmresearchcenter/open_universal_arabic_asr_leaderboard --repo-type space --local-dir open_universal_arabic_asr_leaderboard
cd open_universal_arabic_asr_leaderboard
pip install gradio==5.49.0
pip install -r requirements.txt
python app.py

How to use it: Online, visit the Space on Hugging Face and read the leaderboard that loads. On your own computer, run the start command and open the local page Gradio serves.

python app.py
Tool | What it does
app.py | Starts the Gradio leaderboard
requirements.txt | Lists the Python packages to install
banner.png | The banner picture used by the page

Example: You want an Arabic transcription model for audio in several dialects. You visit the Space, scan the leaderboard of the models it names, and write down the ones worth testing yourself.

For newcomers and experts: If you are new, visit the Space and look at the list of models. ASR means automatic speech recognition, which turns spoken words into text, and the leaderboard helps you see which open models are in the running for Arabic.

If you know the field, the Space names six models, including Sygma-ASR-Arabic, hubert-large-arabic-transcribe, two Audar-ASR-V1 variants and two Cohere transcribe models, plus the Casablanca dataset from UBC-NLP. The trade-off is that a leaderboard reflects one benchmark, so check it against your own audio.

You visit the Gradio page
The app shows the benchmark results
You compare the listed models
You test a shortlist on your audio

Good to know:

  • Scope: it covers open-source Arabic speech recognition models across dialects
  • Hardware: the Space is set to cpu-basic, so no GPU is named
  • License: apache-2
  • Cost to run: none named
  • Data it reaches: the models and the Casablanca dataset it names on Hugging Face
  • Not for: transcribing your own files, since it is a benchmark and not a transcription tool

On Hugging Face, 16 November 2026

  • Likes: 126
  • Built with: gradio
  • License: apache-2.0

Links

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.

Files
README.md  5 KB

1 file, 5 KB, downloaded as a zip.

What’s new

  • Version 1 · 2026-11-16

Compare with similar

Listed Hugging Face Spaces in the same category that share the most tags with this one.

ProductLikesBuilt withLicenseLast change
Open Universal Arabic Asr Leaderboard by elmresearchcenter This page126gradioapache-2.0
Open VLM LeaderboardShares leaderboard · gradio1,035gradioapache-2.0
GAIA LeaderboardShares leaderboard · gradio635gradioapache-2.0
Open VLM Video LeaderboardShares leaderboard · gradio136gradioapache-2.0
Open PL LLM LeaderboardShares leaderboard · gradio84gradioapache-2.0
Arabic LLM Leaderboard - Broad (ABL)Shares leaderboard · gradio58gradioapache-2.0
Numbers as Mindket last read them, from 2026-11-12 to 2026-11-16.

Reviews

No ratings yet

Write a review

No reviews yet.

More from Mindket

See all from Mindket

all-mpnet-base-v2

sentence-transformers · Kit

Turns sentences and paragraphs into 768-number vectors so you can search, cluster and compare text by meaning. It suits developers and data teams working in En…

New

multilingual-e5-base

intfloat · Kit

Turns text in about 100 languages into vectors so you can compare meaning across them, for search and matching. Made for developers who build multilingual retr…

New

sprite-gen

aldegad · Skill

Turns one base drawing of a character into game-ready sprites, either an atlas with a manifest or transparent motion loops, for game developers working in Code…

New4097

Similar in Research

See all in Research

J++ Lenses

koayon · Kit

Shows which words a language model is leaning toward at its middle layers. Lens files for six open models, made for interpretability researchers who want to in…

New

Claude Code skills that guide academic research from planning to publication, with your confirmation at every stage, integrity checks and an optional citation…

New57.1k

GAIA Leaderboard

gaia-benchmark · Kit

Shows a leaderboard built around the GAIA dataset, so people can compare AI systems on one shared benchmark. It is a Gradio app from the gaia-benchmark group.

New