Italian Robust ASR

Italian Robust ASR

Transcribes Italian speech to text with wav2vec2 XLS-R models from a Hugging Face speech challenge. For people who want to try Italian speech recognition in a browser.

Made by dbdmg 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 demo for trying Italian speech recognition models. It came out of a speech challenge run by Hugging Face. The Space names three wav2vec2 XLS-R models, one marked 1b and two marked 300m, and turns Italian audio into text.

Who it suits: It suits people who want to hear how an Italian speech recognizer copes with real recordings, and developers who want a base to test their own models.

Industry | Who uses it | How they use it | Example
Research | Speech researchers | Try Italian recognition models | Run a demo clip and read the text that comes back
Education | Italian learners and teachers | Check spoken Italian against text | Compare what is said with what is written
Media | Editors and journalists | Draft a transcript from audio | Get a first pass to correct by hand
Development | Developers | Test a model before using it in an app | Run a sample clip through the demo

You can skip it if you need languages other than Italian.

Where to use it:

• A web browser, on the Space's page at Hugging Face
• Your own computer, on Windows, macOS or Linux

Use case | Who | What they get back
Transcribe a clip | Researchers | Italian text from speech
Try the demo audio | Newcomers | A quick look at the output using the included clips
Test a model | Developers | A feel for how the models behave

How to set it up:

• Nothing to install if you use the Space online
• Python, if you run a copy on your own computer
• The Python packages in requirements.txt
• The system packages in packages.txt, which Hugging Face installs with apt-get

1. Open the Space's page on Hugging Face to use it online.
2. To run a copy, install huggingface_hub and download the Space into a folder.
3. Move into that folder and install Gradio, then the listed packages.
4. Start the app with Python.

pip install -U huggingface_hub
hf download dbdmg/robust-asr-it --repo-type space --local-dir robust-asr-it
cd robust-asr-it
pip install gradio
pip install -r requirements.txt
python app.py

How to use it: Open the Space and give it Italian audio. The repository holds three short demo clips you can try. On your own copy, start the app with this command and open the local page Gradio prints.

python app.py

Example: Run one of the demo clips from the repository through the app. You get back the Italian text the model hears in the clip, which you can read against the audio.

For newcomers and experts: If you are new, try a demo clip first and read the text it returns. Speech recognition turns spoken words into written ones, and this demo does that for Italian.

If you know the field, the Space lists three Italian wav2vec2 XLS-R checkpoints from the same author, in a plain variant and two others. It is a fork of another Hugging Face speech recognition Space, so the structure is the standard Gradio one. The README gives no accuracy figures.

You give the app Italian audio
A wav2vec2 XLS-R model reads the sound
The app returns the text it heard

Good to know:

• License: mit
• Cost to run: none named. The online Space runs on Hugging Face.
• Data it reaches: the audio you give it, and the models it names on Hugging Face.

On Hugging Face, 13 November 2026

• Likes: 1
• Built with: gradio
• License: mit

Links

• Space: https://huggingface.co/spaces/dbdmg/robust-asr-it
• Files: https://huggingface.co/spaces/dbdmg/robust-asr-it/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.