datasets
Training and evaluation data, with the modality, task and licence stated up front. Listed live from the Hugging Face Hub.
zarma_tts_dataenglish-zarma_sentence-pairs_mt560
English-Zarma Parallel Dataset
This dataset contains parallel sentences in English and Zarma (Niger).
Dataset Information
Language Pair: English ↔ Zarma
Language Code: dje
Country: Niger
Original Source: OPUS MT560 Dataset
Dataset Structure
The dataset contains parallel sentences that can be used for:
Machine translation training
Cross-lingual NLP tasks
Language model fine-tuning
Citation
If you use this dataset, please cite the citation guide of… See the full description on the dataset page: https://huggingface.co/datasets/michsethowusu/english-zarma_sentence-pairs_mt560.ZarmaLanguageRules
Description
This repo contains 20 grammar rules for Zarma language. The rules were create as part of the R2T project which is a proof of concept of a paradigm: Principled Learning (PrL).
Citation
If you use this resource, please use this citation.
@article{r2t2025,
title={R2T: A Case Study in Principled Learning for Low-Resource POS Tagging},
author={Author1 and Author2 and Author3},
journal={TBD},
year={2025},
url={TBD}
}
Acknowledgments
We thank… See the full description on the dataset page: https://huggingface.co/datasets/27Group/ZarmaLanguageRules.noisy_zarma
Zarma Noisy Dataset
Overview
The Zarma Noisy Dataset is a collection of Zarma sentences with artificially introduced noise to simulate human-like errors. This dataset is designed for tasks such as grammatical error correction (GEC), text denoising, and robustness testing in natural language processing (NLP) for low-resource languages like Zarma. It is derived from a clean monolingual Zarma dataset (monolingual_zarma.jsonl) by applying various types of noise… See the full description on the dataset page: https://huggingface.co/datasets/Zauberman/noisy_zarma.noisy_zarma
Zarma Noisy Dataset
Overview
The Zarma Noisy Dataset is a collection of Zarma sentences with artificially introduced noise to simulate human-like errors. This dataset is designed for tasks such as grammatical error correction (GEC), text denoising, and robustness testing in natural language processing (NLP) for low-resource languages like Zarma. It is derived from a clean monolingual Zarma dataset (monolingual_zarma.jsonl) by applying various types of noise, including… See the full description on the dataset page: https://huggingface.co/datasets/27Group/noisy_zarma.bible_zarmaZarma_POS
Zarma_POS Dataset
Overview
The Zarma_POS dataset is a part-of-speech (POS) tagged corpus for the Zarma language, derived from the 27Group/Feriji dataset's fr_dje_corpus subset.
Each entry in the dataset contains:
text: The original Zarma sentence.
tokens: A list of tokenized words and punctuation.
tags: A list of POS tags corresponding to each token (e.g., NOUN, VERB, PUNCT).
Example Entry
{
"text": "Waybora di alboro.",
"tokens": ["Waybora", "di"… See the full description on the dataset page: https://huggingface.co/datasets/27Group/Zarma_POS.zarma-tts-datasetZarma_NER
ZarmaNER-600 Dataset
Dataset Description
ZarmaNER-600 is a gold-standard dataset for Named Entity Recognition (NER) in Zarma. This dataset contains 600 manually annotated sentences, making it the first publicly available NER corpus for Zarma. It was created to support research in low-resource NLP, particularly for sequence tagging tasks, as part of the Rule-to-Tag (R2T) framework introduced in our paper, "R2T: A Case Study in Principled Learning for Low-Resource POS… See the full description on the dataset page: https://huggingface.co/datasets/27Group/Zarma_NER.