Datasets:

Favourez
/

bulu_english

dataset: fr-en-bulu-translation
language:
  - fr
  - en
  - bu
multilinguality: translation
license: mit
task_categories:
  - translation
task_ids:
  - machine-translation

French-English to Bulu Translation Dataset

Dataset Summary

This dataset consists of parallel sentence pairs for machine translation from French and English to Bulu, a low-resource Bantu language spoken in Cameroon. It aims to support the development of machine translation and NLP tools for underrepresented African languages.

Supported Tasks and Leaderboards

Machine Translation: Translate from French or English into Bulu.
Low-resource language modeling: Ideal for training multilingual or translation models for Bulu.

Languages

fr: French
en: English
bu: Bulu (ISO 639-3: bum)

Dataset Structure

The dataset is provided as an Excel file with the following columns:

french_sentence	english_sentence	bulu_sentence
Bonjour	Hello	Mbolè

Each row represents a single translation entry.

Data Splits

All data is currently in one file. Users may split it manually into train/validation/test sets as needed.

Usage

from datasets import load_dataset

dataset = load_dataset("your-username/fr-en-bulu-translation")
print(dataset["train"][0])


@dataset{fr_en_bulu_translation_2025,
  title     = {French-English to Bulu Translation Dataset},
  author    = {Nopole Flairan, Ngana Noa},
  year      = {2025},
  publisher = {Hugging Face},
  note      = {Low-resource language dataset for Bulu translation.}
}