README.md

metadata

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.