Datasets | Mozilla Data Collective

Common Voice

Common Voice Scripted Speech 26.0 - Seediq

A collection of read speech recordings in Seediq.

License: CC0-1.0

Locale: trv

Task: ASR

Format: MP3

Size: 204.97 MB

Taruen

Uzbek Proverbs Text Corpus

A corpus of traditional Uzbek proverbs digitized from a 2005 volume, designed for NLP tasks and LLM cultural competence evaluations.

License: CC0-1.0

Locale: uz

Task: NLP

Format: TXT

Size: 142.68 KB

RFE/RL

RFE/RL Tatar-Bashkir News Text Corpus

This dataset is longitudinal news corpus for the Tatar and Bashkir languages, sourced from Azatliq Radiosi, from 2001 to 2025, it contains over 105,000 articles (30M tokens).

License: CC-BY-NC-SA-4.0

Locale: tt,ba,ru

Task: NLP

Format: TXT

Size: 102.44 MB

Community

Sindhi Word Segmentation (SdSEG)

A sentence-level annotated dataset for training and evaluating sequence labeling models such as CRF, BiLSTM, and transformer-based architectures.

License: CC-BY-NC-4.0

Locale: snd

Task: NLP

Format: JSON

Size: 1.13 MB

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

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

Kazakh Proverbs Text Corpus

A corpus of traditional Kazakh proverbs from the Babalar sözi series (vols 65-69), designed for NLP and LLM cultural competence evaluations.

License: CC0-1.0

Locale: kk

Task: NLP

Format: TXT

Size: 592.14 KB

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

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

Common Voice

Common Voice Scripted Speech 26.0 - Tajik

A collection of read speech recordings in Tajik.

License: CC0-1.0

Locale: tg

Task: ASR

Format: MP3

Size: 19.65 MB

Kaleem Art Press

Kaleem Art Press Saraiki Literature Corpus

This corpus contains approximately one million tokens of Saraiki text curated over the past ten years by Kaleem Art Press. It features a wide range of literary genres, including stories, short stories, novels, fiction, non-fiction, travelogues, poetry, biographies, and historical writings. All content is shared with full author approval. The dataset is intended to support linguistic research, Saraiki language technology development, and the preservation of cultural and literary heritage.

License: CC-BY-NC-4.0

Locale: skr

Task: OTH

Format: TXT

Size: 1.84 MB

RFE/RL

RFE/RL Kazakh News Text Corpus

Longitudinal Kazakh news corpus from Radio Azattyq (2003-2026) with over 139,000 articles and 35M tokens.

License: CC-BY-NC-SA-4.0

Locale: kk

Task: NLP

Format: TXT

Size: 126.81 MB

Digital Divide Data

ddd-kenya-somali-68hrs-asr-part1

A collection of high-quality, crowdsourced audio recordings and verified transcriptions for Somali language, produced by Digital Divide Data

License: CC-BY-4.0

Locale: som

Task: ASR

Format: WAV, TSV

Size: 7.68 GB

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

Community

Sindhi Corpus

SdCorpus is a large scale monolingual Sindhi text corpus collected from publicly available web sources for NLP and language model development.

License: CC-BY-NC-4.0

Locale: snd

Task: LM

Format: TXT

Size: 321.00 MB

RFE/RL

RFE/RL Crimean Tatar News Text Corpus

Longitudinal Crimean Tatar news corpus from Qırım.Aqiqat (2014-2026) with over 32,000 articles and 7.5M tokens.

License: CC-BY-NC-SA-4.0

Locale: crh

Task: NLP

Format: TXT

Size: 18.35 MB

Baloch Publishers Multan

Baloch Publishers Saraiki Literature Corpus

This corpus is a collection of one million tokens of Saraiki language. The data was produced under the Baloch Publishers over the last ten years. The corpus contains work of literature including dictionaries, short stories, novels, fiction, non-fiction, and travelogue. The data is being shared with the approval of the authors. It aims to support linguistic research, language technology development, and cultural preservation.

License: CC-BY-NC-4.0

Locale: skr

Task: NLP

Format: TXT

Size: 2.04 MB

Digital Divide Data

ddd-kenya-somali-68hrs-asr-part2

A collection of high-quality, crowdsourced audio recordings and verified transcriptions for the Somali language, produced by Digital Divide Data.

License: CC-BY-4.0

Locale: som

Task: ASR

Format: WAV, TSV

Size: 8.07 GB

Digital Divide Data

ddd-kenya-somali-68hrs-asr-part3

A collection of high-quality, crowdsourced audio recordings and verified transcriptions for the Somali language, produced by Digital Divide Data.

License: CC-BY-4.0

Locale: som

Task: ASR

Format: WAV, TSV

Size: 1.33 GB

Common Voice

Common Voice Spontaneous Speech 4.0 - Serian Bidayuh

A collection of spontaneous responses to questions in Serian Bidayuh.

License: CC0-1.0

Locale: sdo

Task: ASR

Format: MP3

Size: 201.27 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

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

Common Voice

Common Voice Scripted Speech 26.0 - Saraiki

A collection of read speech recordings in Saraiki.

License: CC0-1.0

Locale: skr

Task: ASR

Format: MP3

Size: 133.49 MB

RFE/RL

RFE/RL Afghan Pashto News Text Corpus

This dataset is a longitudinal news corpus for the Pashto language, sourced from Radio Azadi, from 2000 to 2026, it contains over 204,000 articles (16M tokens).

License: CC-BY-NC-SA-4.0

Locale: ps

Task: NLP

Format: TXT

Size: 117.68 MB

All Datasets

Not sure where to start?

Common Voice Scripted Speech 26.0 - Seediq

Uzbek Proverbs Text Corpus

RFE/RL Tatar-Bashkir News Text Corpus

Sindhi Word Segmentation (SdSEG)

RFE/RL Tajik News Text Corpus

RFE/RL Azerbaijani News Text Corpus

Kazakh Proverbs Text Corpus

Jazab Sindhi Newspaper Corpus

Sindh Sujag Newspaper Corpus

Common Voice Scripted Speech 26.0 - Tajik

Kaleem Art Press Saraiki Literature Corpus

RFE/RL Kazakh News Text Corpus

ddd-kenya-somali-68hrs-asr-part1

RFE/RL Uzbek News Text Corpus

Sindhi Corpus

RFE/RL Crimean Tatar News Text Corpus

Baloch Publishers Saraiki Literature Corpus

ddd-kenya-somali-68hrs-asr-part2

ddd-kenya-somali-68hrs-asr-part3

Common Voice Spontaneous Speech 4.0 - Serian Bidayuh

Tamir Sindhi News Corpus

RFE/RL Turkmen News Text Corpus

Common Voice Scripted Speech 26.0 - Saraiki

RFE/RL Afghan Pashto News Text Corpus