metadata

license: cc-by-4.0
task_categories:
  - translation
language:
  - en
  - nys
tags:
  - machine-translation
  - low-resource
  - african-languages
  - English
  - Giriama
  - agriculture
  - bible
  - parallel-corpus
pretty_name: English-Giriama Parallel Sentence Dataset
size_categories:
  - 1K<n<10K

🗣️ English-Giriama Parallel Sentence Dataset

This dataset consists of sentence pairs in English and their corresponding translations in Giriama (Kigiryama), a Bantu language spoken primarily in coastal Kenya. It supports machine translation (MT) and other cross-lingual NLP tasks, especially in low-resource language research.

📋 Dataset Structure

Each row in the dataset contains:

English Sentence: A sentence in standard English.
Giriama Translation: The corresponding sentence translated into Giriama.

Example:

English Sentence	Giriama Translation
Jesus was born in Judea.	Yesu azhalwa mu Yudea.
The crops need to be watered.	Mihogo chikala kinywe chingonji.

📦 Dataset Details

Language Pair: English ↔ Giriama (Kigiryama)
Number of Sentence Pairs: 7820
Data Sources:
- The Giriama New Testament corpus (verse-aligned, literary/religious domain).
- ~600 domain-specific sentence pairs focused on agriculture and farming contexts. These include vocabulary and expressions commonly used by field officers, farmers, and agri-extension services in rural Kenya.
Format: CSV

🧑‍💻 Intended Use

Machine Translation (MT) — English ↔ Giriama.
Low-resource language research and fine-tuning.
Agricultural tech and rural communication tools.
Cross-lingual representation learning.
Language preservation and documentation.

🚫 Limitations

Religious text data (from the New Testament) may introduce stylistic or archaic tone in parts of the dataset.
The dataset may not cover the full spectrum of informal or modern spoken Giriama.
Agricultural entries were manually curated and may carry contextual assumptions.

🤝 Citation

If you use this dataset in your research or application, please cite it as:

@misc{english_giriama_dataset,
  title = {English-Giriama Parallel Sentence Dataset},
  author = {Lingua-Connect},
  year = {2025},
  howpublished = {\url{https://huggingface.co/datasets/English-Giriama-Dataset}}
}