---
language:
- udm
size_categories:
- 10K<n<100K
dataset_info:
  features:
  - name: rus
    dtype: string
  - name: udm
    dtype: string
  - name: eng
    dtype: string
  splits:
  - name: train
    num_bytes: 18660993
    num_examples: 40115
  download_size: 8913607
  dataset_size: 18660993
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
---

# Udmurt-Russian-English parallel corpora for LaBSE training

English column was translated by Yandex.Translator API from Russian.