datasets
Training and evaluation data, with the modality, task and licence stated up front. Listed live from the Hugging Face Hub.
SOLD
SOLD - A Benchmark for Sinhala Offensive Language Identification
In this repository, we introduce the {S}inhala {O}ffensive {L}anguage {D}ataset (SOLD) and present multiple experiments on this dataset. SOLD is a manually annotated dataset containing 10,000 posts from Twitter annotated as offensive and not offensive at both sentence-level and token-level. SOLD is the largest offensive language dataset compiled for Sinhala. We also introduce SemiSOLD, a larger dataset containing more… See the full description on the dataset page: https://huggingface.co/datasets/sinhala-nlp/SOLD.sinhala-tts-dataset-archive-20260429-082457
Sinhala TTS Dataset
Clean, segmented Sinhala speech from the "Unlimited History" YouTube series by @sunchare.
Stats
Metric
Value
Utterances
218
Train
208
Val
10
Hours
0.51
Mean duration
8.5s
Sample rate
22050 Hz
Pipeline
Raw YouTube audio -> HTDemucs -> VoiceFixer + DeepFilterNet3 ->
Diarization -> Silero-VAD -> ASR (faster-whisper: C:\Users\kosal\sinhala-tts\whisper-small-si-ct2) -> Quality filtering (SNR>=20.0dB)
Format… See the full description on the dataset page: https://huggingface.co/datasets/outlawmold/sinhala-tts-dataset-archive-20260429-082457.named-entity-recognition
Sinhala Named Entity Recognition
Sinhala Named Entity Recognition is a token-level named entity recognition dataset for Sinhala.
This repository is a re-upload of the original Sinhala NER dataset introduced by Manamini et al. (2016) in "Ananya - a Named-Entity-Recognition (NER) System for Sinhala Language" with proper train/ test splits. The dataset was subsequently included as the Named Entity Recognition (NER) task in the SINHALA-GLUE benchmark introduced in "Sinhala… See the full description on the dataset page: https://huggingface.co/datasets/sinhala-nlp/named-entity-recognition.NSINA-Categories
Sinhala News Category Prediction
This is a text classification task created with the NSINA dataset. This dataset is also released with the same license as NSINA.
Data
Data can be loaded into pandas dataframes using the following code.
from datasets import Dataset
from datasets import load_dataset
train = Dataset.to_pandas(load_dataset('sinhala-nlp/NSINA-Categories', split='train'))
test = Dataset.to_pandas(load_dataset('sinhala-nlp/NSINA-Categories', split='test'))… See the full description on the dataset page: https://huggingface.co/datasets/sinhala-nlp/NSINA-Categories.sinhala-sentiment-analysisSinhala-News-Category-classificationThis file contains news texts (sentences) belonging to 5 different news categories (political, business, technology, sports and Entertainment). The original dataset was released by Nisansa de Silva (Sinhala Text Classification: Observations from the Perspective of a Resource Poor Language, 2015). The original dataset is processed and cleaned of single word texts, English only sentences etc.
If you use this dataset, please cite {Nisansa de Silva, Sinhala Text Classification: Observations from… See the full description on the dataset page: https://huggingface.co/datasets/NLPC-UOM/Sinhala-News-Category-classification.pali-sinhalasinhala-english-singlish-translation
Sinhala–English–Singlish Translation Dataset
A parallel corpus of Sinhala sentences, their English translations, and romanized Sinhala (“Singlish”) transliterations.
📋 Table of Contents
Dataset Overview
Installation
Quick Start
Dataset Structure
Usage Examples
Citation
License
Credits
Dataset Overview
Description: 34,500 aligned triplets of
Sinhala (native script)
English (human translation)
Singlish (romanized Sinhala)… See the full description on the dataset page: https://huggingface.co/datasets/Programmer-RD-AI/sinhala-english-singlish-translation.SemiSOLD
SOLD - A Benchmark for Sinhala Offensive Language Identification
In this repository, we introduce the {S}inhala {O}ffensive {L}anguage {D}ataset (SOLD) and present multiple experiments on this dataset. SOLD is a manually annotated dataset containing 10,000 posts from Twitter annotated as offensive and not offensive at both sentence-level and token-level. SOLD is the largest offensive language dataset compiled for Sinhala. We also introduce SemiSOLD, a larger dataset containing more… See the full description on the dataset page: https://huggingface.co/datasets/sinhala-nlp/SemiSOLD.semantic-textual-similarity
Sinhala Semantic Textual Similarity
Sinhala Semantic Textual Similarity is a sentence-pair similarity dataset for Sinhala.
This repository is a re-upload of the original Sinhala Semantic Textual Similarity dataset introduced by Kadupitiya et al. (2016) in "Sinhala Short Sentence Similarity Calculation using Corpus-Based and Knowledge-Based Similarity Measures".
The dataset was subsequently included as the Semantic Textual Similarity (STS) task in the SINHALA-GLUE benchmark… See the full description on the dataset page: https://huggingface.co/datasets/sinhala-nlp/semantic-textual-similarity.Twitter_Sinhala_Hate_SpeechSinhala_dataset_TestSinhala-News-Source-classificationThis dataset contains Sinhala news headlines extracted from 9 news sources (websites) (Sri Lanka Army, Dinamina, GossipLanka, Hiru, ITN, Lankapuwath, NewsLK,
Newsfirst, World Socialist Web Site-Sinhala). This is a processed version of the corpus created by Sachintha, D., Piyarathna, L., Rajitha, C., and Ranathunga, S. (2021). Exploiting parallel corpora to improve multilingual embedding based document and sentence alignment. Single word sentences, invalid characters have been removed from the… See the full description on the dataset page: https://huggingface.co/datasets/NLPC-UOM/Sinhala-News-Source-classification.sinhala-headline-prediction
Sinhala Headline Prediction
Sinhala Headline Prediction is a binary text classification dataset for Sinhala introduced as part of the SINHALA-GLUE benchmark in the ACL 2025 paper "Sinhala Encoder-only Language Models and Evaluation".
Given a Sinhala news article and a candidate headline, the task is to determine whether the headline is the correct headline for the article.
The dataset contains 7,870 training examples and 1,970 test examples.
Dataset Description… See the full description on the dataset page: https://huggingface.co/datasets/sinhala-nlp/sinhala-headline-prediction.CNN-Daily-Mail-Sinhala
Dataset Summary
This dataset card aims to be creating a new dataset or Sinhala news summarization tasks. It has been generated using [https://huggingface.co/datasets/cnn_dailymail] and google translate.
Data Instances
For each instance, there is a string for the article, a string for the highlights, and a string for the id. See the CNN / Daily Mail dataset viewer to explore more examples.
{'id': '0054d6d30dbcad772e20b22771153a2a9cbeaf62',
'article': '(CNN) -- An American… See the full description on the dataset page: https://huggingface.co/datasets/Hamza-Ziyard/CNN-Daily-Mail-Sinhala.BBC-Sinhalasinhala-1.5B-corpusSinhala-1.5B-corpus is a large-scale monolingual corpus for Sinhala, released as part of the ACL 2025 paper "Sinhala Encoder-only Language Models and Evaluation".
The corpus was created to address the limited availability of large-scale pre-training resources for Sinhala. It combines Sinhala text from a range of domains, including web content, news, social media, books, educational material and government documents.
The corpus was used to train the Sinhala encoder-only language models… See the full description on the dataset page: https://huggingface.co/datasets/sinhala-nlp/sinhala-1.5B-corpus.Sinhala-Script-LangID-Benchmarksinhala_sa
Sentiment Analysis Data for the Sinhala Language
Dataset Description:
This dataset contains a sentiment analysis data from Ranathunga et al (2021).
Data Structure:
The data was used for the project on injecting external commonsense knowledge into multilingual Large Language Models.
Citation:
@article{ranathunga2021sentiment,
title={Sentiment analysis of sinhala news comments},
author={Ranathunga, Surangika and Liyanage, Isuru Udara},
journal={Transactions on Asian and… See the full description on the dataset page: https://huggingface.co/datasets/DGurgurov/sinhala_sa.sinhala-comment-popularity-predictionsinhala-summarization-dataset
Sinhala Text Summarization Dataset
Dataset Description
This dataset is a Sinhala text summarization dataset created for research in low-resource language summarization. The dataset contains 2,493 Sinhala article-summary pairs collected from diverse publicly accessible Sinhala online sources.
This repository contains a Sinhala article-summary dataset introduced in the following IEEE conference publication:
Sinhala Automatic Text Summarization: Dataset Creation and… See the full description on the dataset page: https://huggingface.co/datasets/hans1k/sinhala-summarization-dataset.sinhala_conceptnet
ConceptNet Data for the Sinhala Language
Dataset Description:
This dataset contains data extracted from ConceptNet using the dedicated module for fetching knowledge from the graph, available on GitHub.
Data Structure:
The data is converted from triplets into natural text using a pre-defined relationship mapping and split into training and validation sets. It was used for training language adapters for the project aimed at injecting external commonsense knowledge into multilingual… See the full description on the dataset page: https://huggingface.co/datasets/DGurgurov/sinhala_conceptnet.sinhalaFacebook_Sinhala_Hate_Speechsinhala-articles
Sinhala Articles Dataset
A large-scale, high-quality Sinhala text corpus curated from diverse sources including news articles, Wikipedia entries, and general web content. This dataset is designed to support a wide range of Sinhala Natural Language Processing (NLP) tasks.
📊 Dataset Overview
Name: Navanjana/sinhala-articles
Total Samples: 2,148,688
Languages: Sinhala (si)
Features:
text: A single column containing Sinhala text passages.
Size: Approximately 1M < n <… See the full description on the dataset page: https://huggingface.co/datasets/Navanjana/sinhala-articles.Sinhala-Dataset-Questions-and-Answers
Sinhalese Q&A Dataset
Dataset Description
This dataset consists of question-and-answer pairs in Sinhalese (සිංහල) language.Each row contains a question in Sinhalese and its corresponding answer, also in Sinhalese.The dataset is intended for training and evaluating natural language processing models on tasks such as question answering, dialogue systems, and educational tools.
Key facts
Language(s): Sinhalese (si)
Size: ~ examples
License:
Usage domain:… See the full description on the dataset page: https://huggingface.co/datasets/ChamaraVishwajithRajapaksha/Sinhala-Dataset-Questions-and-Answers.sinhala-spell-correction-dataset
Sinhala Spell Correction Dataset
A Sinhala spell correction dataset used for training and evaluating neural spell correction models as part of the LMSpell project.
Dataset Description
This dataset combines data from previously published Sinhala spell correction resources and applies additional cleaning to improve its suitability for training neural spell correction models.
The dataset originates from the benchmark introduced by Sonnadara et al. (2021) and was… See the full description on the dataset page: https://huggingface.co/datasets/lm-spell/sinhala-spell-correction-dataset.HelaTransformerSinhala-Corpussinmin