SalomonMetre13/cd_multi_parallel_31k_v1

Dataset Summary

cd_multi_parallel_31k_v1 is a mid-scale multilingual parallel corpus for low-resourced languages of the Democratic Republic of the Congo (DRC). It consists of 31,440 aligned sentences, enabling research in machine translation, multilingual NLP, and linguistic analysis. This dataset is extracted from publicly available Bible data, providing structured and well-aligned multilingual text.

Features

• 🌍 Multilinguality: Includes translations in 15 Congolese languages along with 6 high-resource international languages.
• 🔗 Parallel Data: Sentences are aligned across languages, making it suitable for training and evaluating machine translation models.
• ✅ High Quality: Extracted from publicly available Bible data and carefully processed to maintain alignment integrity.
• 🔄 Versioning: This is version 1.

Languages Included

The dataset includes 15 Congolese languages and 6 widely spoken international languages, making it valuable for both local and global NLP research.

Dataset Structure

Each entry in the dataset contains:

• 🆔 A unique sentence ID
• 🌐 Translations in multiple languages

Usage

📥 Loading the dataset

You can load the dataset using the Hugging Face datasets library:

from datasets import load_dataset

dataset = load_dataset("SalomonMetre13/cd_multi_parallel_31k_v1")