|
|
--- |
|
|
license: cc-by-4.0 |
|
|
language: |
|
|
- crh |
|
|
- en |
|
|
- ru |
|
|
- uk |
|
|
task_categories: |
|
|
- translation |
|
|
tags: |
|
|
- crh |
|
|
- parallel-corpus |
|
|
- mt |
|
|
- Crimean Tatar |
|
|
- Machine Translation |
|
|
pretty_name: Crimean Tatar-English/Russian/Ukrainian Parallel Corpora |
|
|
configs: |
|
|
- config_name: default |
|
|
data_files: |
|
|
- split: crh_eng |
|
|
path: eng_crh* |
|
|
- split: crh_rus |
|
|
path: rus_crh* |
|
|
- split: crh_ukr |
|
|
path: ukr_crh* |
|
|
size_categories: |
|
|
- 100K<n<1M |
|
|
--- |
|
|
|
|
|
# Crimean Tatar-English/Russian/Ukrainian Parallel Corpora |
|
|
|
|
|
## Overview |
|
|
This repository contains the Crimean Tatar-English/Russian/Ukrainian parallel corpora, a collection of sentences in Crimean Tatar and their corresponding translations in English/Russian/Ukrainian. The dataset is intended for use in natural language processing (NLP) tasks such as machine translation and cross-lingual analysis. |
|
|
All sentences in Crimean Tatar are written in Cyrillic and/or Latin script. If any of the transliterations are missing, they are skipped in the dataset. You can transliterate between Cyrillic and Latin using [this transliterator](https://github.com/qirimca/Transliterator) |
|
|
|
|
|
- **Dataset Name:** Crimean Tatar-English/Russian/Ukrainian Parallel Corpora |
|
|
- **Languages:** Crimean Tatar, English, Russian, Ukrainian |
|
|
- **Domain:** General (Mostly books) |
|
|
|
|
|
```python |
|
|
from datasets import load_dataset |
|
|
|
|
|
# Load the Crimean Tatar-English/Russian/Ukrainian parallel corpora |
|
|
dataset = load_dataset('QIRIM/crh-parallel-corpora') |
