Belarusian datasets

Belarusian datasets

A browsable list of hand-picked datasets and corpora for the Belarusian language, covering speech, text, parallel texts, benchmarks, dictionaries and OCR. The page is in English and Belarusian.

Made by Zhnivo 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 static web page that lists datasets and corpora made for the Belarusian language. It covers speech, text, parallel texts, benchmarks, dictionaries and OCR, and adds sources from outside Hugging Face. Download counts on the page come live from the Hugging Face API.

Who it suits: People who work with Belarusian and need to find data fast, without searching the whole Hugging Face catalogue. The page shows an English and a Belarusian interface.

Industry | Who uses it | How they use it | Example
Speech technology | Speech engineers | Look for Belarusian audio corpora for recognition work | Finding a speech corpus to train or test a model
Translation | Machine translation developers | Look for parallel texts in Belarusian | Choosing a text pair set for a translation experiment
Linguistics | Language researchers | Find dictionaries, grammar data and text corpora | Locating a grammar database for a study
Document scanning | OCR developers | Find line images and benchmarks for Belarusian print | Picking a synthetic print set for testing

If you do not work with Belarusian text or speech, you can skip it.

Where to use it:

• A web browser, on the Space's page at Hugging Face
• Windows, macOS or Linux, by opening a downloaded index.html in any browser

Use case | Who | What they get back
Find Belarusian datasets | Researchers and developers | A hand-picked list with live download counts
Find sources outside Hugging Face | Anyone collecting data | A separate tab of curated outside sources, such as Common Voice

How to set it up: Online use needs nothing, since the page runs in the browser. To keep a copy, you need little.

• Python and pip, only to download the files with the Hugging Face tool
• A web browser to open the page

1. Install the Hugging Face hub package.
2. Download the Space's files into a local folder.
3. Move into that folder.
4. Open index.html in a web browser. No build step is needed.

pip install -U huggingface_hub
hf download Zhnivo/belarusian-datasets --repo-type space --local-dir belarusian-datasets
cd belarusian-datasets

How to use it: Open the page, then browse the hand-picked sets by kind and read the live download counts. The files in the repository each have a role.

Tool | What it does
index.html, style.css, app.js | Make up the static page itself
data/curated.json | Holds the hand-curated list of what the page shows, plus reviewed and dropped items
data/extra.json | Holds curated sources outside Hugging Face for the Other sources tab
data/hf_snapshot.json | A fallback copy of the Hugging Face numbers for when the live request fails
build_extra.py | Refreshes the snapshot, and its candidates flag prints new items not yet reviewed

Example: You open the page looking for Belarusian speech data. The page shows the hand-picked sets for speech, each with its download count, so you can pick one to open on Hugging Face.

For newcomers and experts: If you are new to this, open the Space's page and scroll through the list first. Nothing needs installing, and you will see which kinds of Belarusian data exist.

If you know the field, the page is plain HTML, CSS and JavaScript with no build step. It shows only what the curated allowlist in data/curated.json contains. Live download counts come from the Hugging Face API, and a saved snapshot takes over when that request fails.

Open the page in a browser
The page reads the curated list in data/curated.json
It asks the Hugging Face API for live counts
On failure it uses the saved snapshot
You browse the datasets

Good to know: The page is a catalogue of hand-picked entries, so it does not list everything that exists for Belarusian.

• License: MIT, as the Space's license line says
• Cost to run: none named
• Data it reaches: the Hugging Face API for live download counts, with a local snapshot as backup

On Hugging Face, 13 November 2026

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

Links

• Space: https://huggingface.co/spaces/Zhnivo/belarusian-datasets
• Files: https://huggingface.co/spaces/Zhnivo/belarusian-datasets/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.