Datasets | Mozilla Data Collective

All Datasets

Compensated Datasets

API

Contribute Datasets

Loading datasets...

Terms

FAQs

Participation Guidelines

Responsible Disclosure

News and Documentation

Data and Model Competitions

[email protected]

Terms

FAQs

Participation Guidelines

Responsible Disclosure

News and Documentation

Data and Model Competitions

[email protected]

Mozilla Data Collective is a mission-locked British company, incubated by Mozilla Foundation, and backed by the not-for-profit Mozilla.org

All Datasets

Not sure where to start?

Describe what you are building and the Data Assistant will find the right datasets for you.

Try the Data Assistant

Filtered by

Search results for “Sylheti Text Corpus”

100 Datasets

MDC Curators

CoVoST 2 Turkish - English

The dataset includes audio files in Turkish and their translations in English.

License: CC-BY-NC-4.0

Locale: tr, en

Task: MT

Format: MP3, TSV

Size: 207.40 MB

Sindh Sujag Newspaper Agency

Sindh Sujag Newspaper Corpus

The corpus contains approximately 1.2 million tokens from the Sindh Sujag Newspaper Agency published between 2024 and 2025. It includes complete newspaper content such as headlines, editorials, finance news, and advertisements. The newspaper is published daily in Karachi, Pakistan, and represents a comprehensive collection of contemporary Sindhi journalistic writing.

License: CC-BY-4.0

Locale: snd

Task: NLP

Format: TXT

Size: 2.63 MB

Institute of African Digital Humanities

Teke-Laali-TTS-Dataset

The dataset contains paired audio and text resources for Teke-Laali, a Bantu language spoken in the Congo. It consists of seven folders containing a total of 9,069 audio clips from raw audio recordings, with a total duration of 7:01:50.126 (HH:MM:SS.mmm). Additionally, there are seven audio/text mapping files containing a total of 9,069 lines. The dataset is suitable for TTS tasks.

License: NOODL-1.0

Locale: lli

Task: TTS

Format: WAV, TSV

Size: 635.61 MB

Institute of African Digital Humanities

Bamun-French Parallel Corpus 1.1

This dataset is an updated version of the "Bamun-French Parallel Corpus", a parallel corpus of texts in Bamun (Shupament) and French.

License: NOODL-1.0

Locale: bax

Task: MT

Format: TSV

Size: 99.78 KB

MEDIAMEN

English–Punjabi (Shahmukhi) Parallel Sentences Corpus (Mediamen Archives)

This parallel sentences corpus containing 30,405 aligned sentence pairs with a total of approximately 0.62 million tokens, curated from the archival materials of Mediamen (Advertising Agency). The sentences were professionally translated from English into Punjabi (Shahmukhi) and are intended to support machine translation, linguistic research, and Punjabi language technology development, particularly for real-world and contemporary language use.

License: CC-BY-NC-4.0

Locale: en-PK, pnb

Task: MT

Format: CSV

Size: 1.08 MB

MDC Curators

CoVoST 2 English - Turkish

The dataset includes audio files in English and their translations in Turkish.

License: CC-BY-NC-4.0

Locale: en, tr

Task: MT

Format: MP3, TSV

Size: 12.56 GB

Jazab Publishers

Jazab Sindhi Newspaper Corpus

The corpus contains 1.07 million tokens from the Jazab a Sindhi Newspaper published from the year 2023-2025. The text consists of the complete newspaper content including headlines, editorials, finance news and advertisements. The newspaper published in Karachi, Pakistan on daily basis.

License: CC-BY-NC-SA-4.0

Locale: snd

Task: NLP

Format: TXT

Size: 2.33 MB

Collaborative Action For Research & Development (CARD)

Gawri (گاؤری) Magazine Corpus

The Gawri (گاؤری) Magazine Corpus is a curated collection of monthly Gawri magazine text (~67,724 tokens) that supports research and language technology.

License: CC-BY-NC-4.0

Locale: gwc

Task: NLP

Format: TXT

Size: 146.71 KB

RFE/RL

RFE/RL Tajik News Text Corpus

Longitudinal Tajik & Russian news corpus from Radio Ozodi (2000-2026) with over 166,000 articles and 20M tokens.

License: CC-BY-NC-SA-4.0

Locale: tg,ru

Task: NLP

Format: TXT

Size: 145.27 MB

Institute of African Digital Humanities

Bati-MultiDialectalASR-Dataset

This dataset contains paired audio and text resources for three Bati dialects (Kelleng, Mbougue, and Nyambat), which belong to the Yambasa group of Bantu languages found in Cameroon. It contains 13,344 audio clips totalling 6 hours, 8 minutes and 12.286 seconds and 44 audio/text mapping files totalling 13,346 lines. Due to its cross-dialectal nature, the dataset is suitable for multilingual automatic speech recognition tasks.

License: NOODL-1.0

Locale: btc

Task: ASR

Format: WAV, TSV

Size: 3.27 GB

Community

Biomedical Urdu Named Entity Recognition (BioUNER)

Biomedical Urdu Named Entity Recognition dataset for advancing healthcare and clinical NLP research in Urdu.

License: CC-BY-NC-4.0

Locale: urd

Task: NLP

Format: CSV

Size: 336.69 KB

RFE/RL

RFE/RL Uzbek News Text Corpus

Longitudinal Uzbek news corpus from Radio Ozodlik (2002-2026) with over 166,000 articles and 31M tokens.

License: CC-BY-NC-SA-4.0

Locale: uz

Task: NLP

Format: TXT

Size: 154.21 MB

RFE/RL

RFE/RL Azerbaijani News Text Corpus

Longitudinal Azerbaijani & Russian news corpus from Radio Azadlıq (2005-2026) with over 239,000 articles and 37M tokens.

License: CC-BY-NC-SA-4.0

Locale: az,ru

Task: NLP

Format: TXT

Size: 211.65 MB

Taruen

Kumyk Proverbs and Sayings Text Corpus

A corpus of traditional Kumyk proverbs and sayings extracted from a 2015 dictionary, designed for NLP tasks and LLM cultural competence evaluations.

License: CC0-1.0

Locale: kum

Task: NLP

Format: TXT

Size: 86.48 KB

MDC Community Concierge

VoxForge - Turkish

3 hours (1717 utterances) of read speech of Türkçe (Turkish).

License: GPL-3.0

Locale: tr

Task: ASR

Format: WAV, TSV

Size: 692.29 MB

Tamir News Agency

Tamir Sindhi News Corpus

The corpus contains 1.1 million tokens from the Tamir Sindhi Newspaper published from the year 2022-2025. The text consists of the complete newspaper content including headlines, editorials, finance news and advertisements. The newspaper published in Karachi, Pakistan on daily basis.

License: CC-BY-NC-SA-4.0

Locale: snd

Task: NLP

Format: TXT

Size: 2.56 MB

PROXIMA

Urdu Medical SFT

Safety aligned Urdu medical question answer dataset of 3,072 records for supervised fine-tuning of medical-domain LLMs.

License: CC-BY-SA-4.0

Locale: urd

Task: NLG

Format: PARQUET

Size: 357.97 KB

Open Home Foundation

Berta 1.0

Text to speech dataset for Hungarian, female speaker, approximately 1 hour of read speech.

License: CC0-1.0

Locale: hu-HU

Task: TTS

Format: FLAC

Size: 209.52 MB

RFE/RL

RFE/RL Turkmen News Text Corpus

Longitudinal Turkmen & Russian news corpus from Azatlyk Radiosy (2009-2026) with nearly 65,000 articles and 16.5M tokens.

License: CC-BY-NC-SA-4.0

Locale: tk,ru

Task: NLP

Format: TXT

Size: 48.28 MB

Institute of African Digital Humanities

Yaka-TTS-Dataset

Paired audio and text data on Yaka (also known as West Teke), a language spoken in Congo. The audio corpus consists of 7,648 clips read by one speaker for a total duration of 344 min 40.48 sec. The dataset also contains a mapping file of audio and text with 7,648 lines. Each line begins with the name of an audio file, followed by a tab and then the corresponding text excerpt. This dataset is suitable for TTS tasks.

License: NOODL-1.0

Locale: iyx

Task: TTS

Format: WAV, TSV

Size: 1.26 GB

MirasAICompensated

🚧 Noakhali 10 Hours TTS (Male Speaker) 🚧

A high-quality Noakhali TTS dataset with paired speech recordings and text transcriptions for speech synthesis, linguistic research, and language preservation.

License: MirasAI Data License

Locale: oak

Task: TTS

Format: WAV, TSV

Size: 2.87 GB

Community

Malayalam Time-Aligned Speech Corpus

A Malayalam speech dataset containing 100 audio files with time-aligned .srt transcriptions from 5 speakers.

License: CC-BY-NC-4.0

Locale: mal

Task: ASR

Format: WAV, SRT

Size: 1.50 GB

Community

Mandar Balanipa Speech Corpus

This Mandar language dataset is text and audio data paired for languange documentation and linguistic research in technology.

License: CC-BY-NC-4.0

Locale: mdr

Task: TTS

Format: WEBM, TSV

Size: 281.70 MB

MirasAI

Saraiki 10 Hours TTS Dataset

A 10-hour Saraiki text-to-speech dataset consisting of recorded speech and aligned transcripts, designed for speech synthesis research and development.

License: CC-BY-NC-SA-4.0

Locale: srk

Task: TTS

Format: WEBM, TSV

Size: 584.44 MB