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

1,200 Datasets

Community

Visual Speech Recognition-Audiovisual Indonesian Dataset (VSR-Aid)

This dataset consists of audio, text, and video of lip movement for about 6 hours.

License: CC-BY-SA-4.0

Locale: ind

Task: OTH

Format: WEBM, TSV

Size: 1.03 GB

Community

Indonesia Lip-Record Speech Dataset

This dataset consists of lip-only video recordings paired with audio and transcribed text in Indonesian.

License: CC-BY-NC-SA-4.0

Locale: ind

Task: OTH

Format: WEBM, TSV

Size: 897.43 MB

MirasAI

Palula Literature Corpus

A literary corpus of Palula texts preserving the language’s rich cultural heritage and supporting research in linguistics, NLP, and language preservation.

License: CC-BY-NC-4.0

Locale: phl

Task: TTS

Format: DOCX, TXT, PDF

Size: 34.05 MB

Community

Indonesian Acculturation Culture - Image Corpus

This dataset contains photographs of cultural objects collected from various locations in East Java and Yogyakarta Province, Indonesia.

License: CC-BY-NC-SA-4.0

Locale: ind

Task: CV

Format: JPEG, WEBM, JSON

Size: 316.47 MB

MDC DatasetsCompensated

🚧 Khowar-English Parallel Corpus 🚧

A parallel corpus of over 10,000 sentences of Khowar and English. The sentences are derived from a multimodal corpus consisting of 152 transcribed and translated Khowar language videos.

License: MDC DLA-1.0

Locale: khw, en

Task: MT

Format: TSV

Size: 407.57 KB

MirasAI

Wakhi Literature Corpus

A literature corpus of Wakhi language texts, preserving the language’s literary, cultural, and linguistic heritage for research and low-resource NLP.

License: CC-BY-ND-4.0

Locale: wbl

Task: NLP

Format: DOCX, TXT

Size: 74.83 KB

MirasAI

Ormuri Bilingual Dictionary by Rozi Khan Burki

A Ormuri dictionary by Rozi Khan Burki documenting Ormuri vocabulary and meanings to support language preservation, linguistic research, and NLP development.

License: CC-BY-ND-4.0

Locale: oru

Task: TTS

Format: JPG, PDF, DOCX

Size: 15.68 MB

Community

Indonesian-Multidialect Lip-Record Speech Corpus

This dataset contains Indonesian speech data designed for lip-reading research and development.

License: CC-BY-SA-4.0

Locale: ind

Task: OTH

Format: WEBM, TSV

Size: 981.20 MB

Content X LabsCompensated

🚧 Thai Episodic Television // Rights-Cleared · 100 Hours 🚧

100 hours of rights-cleared Thai-language episodic television content (5–6 episodes/series), with English subtitles and structured metadata

License: CXL-DLA-1.0

Locale: TH

Task: N/A

Format: MP4, SRT, CSV

Size: 630.79 GB

Community

IND-Jabodetabek Signage and Public Places

This dataset contains annotated images of signs and signboards across the Greater Jakarta (Jabodetabek) area.

License: CC-BY-NC-SA-4.0

Locale: ind

Task: CV

Format: JPEG, WEBM, JSON

Size: 1.56 GB

Community

INALVS (Indonesian Lip-reading & Visual Speech Dataset)

INALVS is an Indonesian visual speech dataset of video recordings for research in lip-reading, visual speech recognition, and artificial intelligence.

License: CC-BY-NC-SA-4.0

Locale: ind

Task: OTH

Format: WEBM, TSV

Size: 1.22 GB

MirasAI

Ormuri Literature corpus by Rozi Burki

Ormuri Literature Corpus is a collection of Ormuri literary texts by Rozi Burki, supporting research and preservation of this under-resourced language.

License: CC-BY-NC-SA-4.0

Locale: oru

Task: NLP

Format: PDF, DOCX

Size: 18.92 MB

Institute of African Digital Humanities

A Tupuri Sociocultural Dataset

Sociocultural dataset for the Tupuri community of Cameroon: identity, kinship, religion, economy, onomastics and a linguistic component.

License: NOODL-1.0

Locale: tui

Task: OTH

Format: TSV, PDF

Size: 357.68 KB

Institute of African Digital Humanities

A Bafia Sociocultural Dataset

Sociocultural dataset for the Bafia community of Cameroon: identity, kinship, religion, economy, onomastics and a linguistic component.

License: NOODL-1.0

Locale: ksf

Task: OTH

Format: TSV, PDF

Size: 358.38 KB

Institute of African Digital Humanities

A Fong Sociocultural Dataset

Sociocultural dataset for the Fong community of Cameroon: identity, kinship, religion, economy, onomastics and a linguistic component.

License: NOODL-1.0

Locale: ewo

Task: OTH

Format: TSV, PDF

Size: 354.12 KB

MirasAI

Dameli Literature Corpus

A literary corpus of Dameli texts preserving the language’s literary traditions, cultural heritage, narratives, and diverse forms of written expression.

License: CC-BY-NC-4.0

Locale: dml

Task: NLP

Format: TXT, DOCX

Size: 9.83 MB

MirasAI

Associated Press Pakistan 21.4 Million Pashto Corpus

Pashto news corpus from the Associated Press of Pakistan, containing 99,495 articles and 21.4 million words from 2015–2026.

License: CC-BY-NC-SA-4.0

Locale: pus

Task: NLP

Format: TXT

Size: 38.81 MB

MirasAI

Associated Press Pakistan 1.8 Million Chinese Corpus

A Chinese-language corpus of about 1.8 million Associated Press Pakistan news articles, useful for NLP and language model training.

License: CC-BY-NC-SA-4.0

Locale: zho

Task: NLP

Format: TXT

Size: 2.76 MB

MirasAI

Associated Press Pakistan 5.1 Million Arabic Corpus

A 5.1-million-word Arabic-language corpus of 41,963 Associated Press Pakistan (APP) news articles (2016–2026).

License: CC-BY-NC-SA-4.0

Locale: arb

Task: NLP

Format: TXT

Size: 9.60 MB

MirasAI

Associated Press Pakistan 78.1 Million English Corpus

The Associated Press of Pakistan 78.1 Million English Corpus is a large-scale English text released for non-commercial research with APP's permission.

License: CC-BY-NC-SA-4.0

Locale: eng

Task: NLP

Format: TXT

Size: 178.01 MB

MirasAI

Associated Press Pakistan 10.4 Million Saraiki Corpus

A 10.4-million-word Saraiki news corpus from Associated Press of Pakistan, covering 2016–2026 for non-commercial Saraiki NLP research.

License: CC-BY-NC-SA-4.0

Locale: skr

Task: NLP

Format: TXT

Size: 19.51 MB

MirasAI

Associated Press Pakistan 74.4 Million Urdu Corpus

A 74.4M-word, 267,887-article Urdu news corpus from Associated Press Pakistan (2015-2026), released for non-commercial research with APP permission.

License: CC-BY-NC-SA-4.0

Locale: urd

Task: NLP

Format: TXT

Size: 143.39 MB

MirasAI

Associated Press Pakistan 16.4 Million Sindhi Corpus

A 16.4-million-word Sindhi-language corpus of 58,815 Associated Press Pakistan (APP) news articles (2015–2026).

License: CC-BY-NC-SA-4.0

Locale: snd

Task: NLP

Format: TXT

Size: 23.23 MB

MirasAI

Associated Press Pakistan 16.9 Million Balochi Corpus

A 16.9M-word Balochi news corpus from Associated Press of Pakistan, covering 59,172 articles published between 2016 and 2026.

License: CC-BY-NC-SA-4.0

Locale: bal

Task: NLP

Format: TXT

Size: 22.85 MB