Loading datasets...
Release Date: 9/2/2026
Format: TXT
Size: 19.51 MB
Share
The Associated Press of Pakistan 10.4 Million Saraiki Corpus is a large-scale collection of Saraiki-language news articles published by the Associated Press of Pakistan. The corpus contains approximately 10.4 million words across 52,388 articles, covering publications from 2016 to 2026. It represents one of the larger documented collections of contemporary Saraiki news text and provides valuable resources for research on the Saraiki language. The corpus was collected from Associated Press of Pakistan's publicly published Saraiki-language news section on app.com.pk using an automated web-crawling pipeline developed by MirasAI. Articles were processed to extract relevant textual and metadata fields, including headlines, article bodies, publication dates, and source URLs. The collected material was deduplicated, cleaned to remove web-page boilerplate, and normalized into plain UTF-8 text while preserving the original Saraiki script.
Licensing
Creative Commons Attribution Non Commercial Share Alike 4.0 International (CC-BY-NC-SA-4.0)
https://spdx.org/licenses/CC-BY-NC-SA-4.0.htmlRestrictions/Special Constraints
This dataset is restricted to approved, non-commercial research use with the Associated Press of Pakistan’s permission. The Associated Press of Pakistan retains full ownership; sale, sublicensing, commercial use, and redistribution are prohibited.
Forbidden Usage
The dataset must not be used for commercial purposes, resale, sublicensing, redistribution, or monetization. It may not be used to develop commercial AI models, products, or services without explicit written permission from the Associated Press of Pakistan.
Ethical Review
Hosted with the express permission of the Associated Press of Pakistan, which retains full ownership. MirasAI acts solely as a documentation and hosting facilitator. Access is restricted and approval-based through the Mozilla Data Collective for non-commercial research only. Sale, sublicensing, and commercial redistribution are prohibited.
Intended Use
This dataset is intended for non-commercial research in Saraiki NLP, computational linguistics, and language technology, including language modeling, corpus analysis, text classification, machine translation, and linguistic research. All use must comply with the access restrictions and Associated Press of Pakistan's terms.
The Associated Press Pakistan (APP) 10.4 Million Saraiki Corpus is a large-scale collection of Saraiki-language news articles published by the Associated Press of Pakistan, Pakistan’s national news agency. Established in 1947, the Associated Press of Pakistan is the country’s premier news service and publishes news across multiple languages, including English, Urdu, Chinese, Arabic, Sindhi, Saraiki, Pashto, Balochi, and Brahvi.
This dataset specifically contains Saraiki-language text and comprises approximately 10.4 million words across 52,388 articles, covering publications from 2016 to 2026. It represents one of the larger documented collections of contemporary written Saraiki news content and provides a valuable resource for research on Saraiki, an under-resourced language in natural language processing (NLP).
Saraiki (also spelled Siraiki) is an Indo-Aryan language spoken primarily in the Saraiki-speaking belt of southern Punjab, Pakistan, with additional speaker communities in Sindh, Balochistan, and the diaspora. It is one of Pakistan's major regional languages and carries a distinct literary and cultural tradition. Saraiki is used in education, media, literature, and everyday communication among Saraiki-speaking communities, and it plays an important role in the development of speech and language technologies for low-resource languages.
| Field | Details |
|---|---|
| Articles | 52,388 |
| Word Count | 10.4 million |
| Coverage | 2016–2026 |
| Language | Saraiki |
| ISO 639-3 Code | skr |
| Script | Perso-Arabic / Shahmukhi |
| Source | Associated Press of Pakistan (APP), Saraiki section |
Articles were collected from APP's publicly published Saraiki-language section on app.com.pk using an automated web crawler developed by MirasAI. Each article's HTML was parsed to extract the headline, body text, publication date, and source URL.
The collected articles were deduplicated, cleaned to remove web-page boilerplate, and normalized to plain text while preserving the original Saraiki script in UTF-8. Crawling was rate-limited and conducted with the knowledge and permission of the Associated Press of Pakistan.
└── Saraki-202608T08354Z-1-001
└── Saraiki
├── 2016.txt
├── 2017.txt
├── 2018.txt
├── 2019.txt
├── 2020.txt
├── 2021.txt
├── 2022.txt
├── 2023.txt
├── 2024.txt
├── 2025.txt
└── 2026.txt ...
اسلام آباد ۔ 2 جنوری (اے پی پی) مسلم لیگ (ن) دے رہنماتے سینیٹر لیفٹیننٹ جنرل (ر) عبدالقیوم آکھیا جو``` پاکستان 2016 اچ مشکلات دے باوجود سیاسی،معاشی، اقتصادی اتے معاشرتی میدان اچ کامیابیاں حاصل کیتیاں ہن۔ ریڈیو پاکستان نال گالھ مہاڑ کریندے ہوئیں انہاں آکھیا جو پاکستان مسلم لیگ ن دے برسر اقتدار امدے ہی ملک اچ دہشت گردی تے توانائی سمیت مختلف بحراناں اچ جکڑیا ہویا ہا تے موجودہ حکومت دیاں پالیسیاں دی بدولت 2016اچ اوکھ دے با وجودسیاسی،معاشی، اقتصادی اتے معاشرتی میدان اچ لاتعدادکامیابیاں حاصل تھیاں ۔انھاں آکھیا جو گزریل دورا چ ملک توانائی بحران اچ جکڑیا ہویا ہا ایندے با وجود صنعتاں کوں بلا تعطل بجلی ڈتی گی وزیر اعظم محمد نواز شریف دیاں کاوشاں نال نویں توانائی پراجیکٹس دا اجراء کیتا گیا جیں نال لوڈشیڈنگ اچ کافی حد تئیں گھٹکی آئی ہے تے اگلے سال تئیں لوڈشیڈنگ تے پورا قابو پا گھدا ویسی ۔
This corpus is uploaded and hosted with the express permission of the Associated Press of Pakistan. The Associated Press of Pakistan retains full and exclusive ownership of the data at all times. MirasAI holds no ownership, copyright, or other proprietary interest and acts solely as documentation and hosting facilitator. The corpus is available under a restricted, approval-based access model on the Mozilla Data Collective access requests are reviewed before release and is licensed for non-commercial research use only. No sale, sublicensing, or commercial redistribution is permitted.
Intended for non-commercial NLP and AI research, including language modeling, machine translation, and information retrieval. All uses must credit the Associated Press of Pakistan as the sole and primary source of the data.