---
language:
- ak
tags:
- akan
- twi
- bible
- text-corpus
---

# Akan Bible Text Corpus

This dataset contains text from the Akan (Twi) Bible, preprocessed and cleaned for NLP tasks such as training language models.

## Dataset Structure

- `train`: 90% of the data
- `test`: 10% of the data

Each row contains a single column:
- `text`: A sentence or merged sentence group from the Twi Bible

## Source

This dataset was scraped from the JWT Twi Bible found here: https://www.jw.org/tw/nhomakorabea/bible/nwt/nhoma/

## Usage

```python
from datasets import load_dataset

dataset = load_dataset("prince4332/akan-bible-text-corpus")
```