Datasets | Mozilla Data Collective

MDC DatasetsCompensated

🚧 Hazargi Multimodal Dataset 🚧

A Hazargi multimodal dataset with 10 hours of synchronized video, audio, transcriptions, speech segmentation, and ELAN annotations.

License: MDC DLA-1.0

Locale: haz

Task: NLP

Format: MP4, EAF

Size: 44.00 GB

MDC DatasetsCompensated

🚧 Hazargi Corpus for Speech Recognition 🚧

A Hazargi ASR dataset with approximately 10 hours of audio with transcriptions.

License: MDC DLA-1.0

Locale: haz

Task: ASR

Format: MP3, TSV

Size: 511.56 MB

MDC DatasetsCompensated

🚧 Northern Pashto Multimodal Dataset 🚧

A Northern Pashto multimodal dataset with 10 hours of synchronized video, audio, transcriptions, speech segmentation, and ELAN annotations.

License: MDC DLA-1.0

Locale: pbu

Task: NLP

Format: MP4, WAV, EAF

Size: 251.41 GB

MDC DatasetsCompensated

🚧 Hazargi-English Parallel Text Corpus 🚧

A Hazargi-English parallel corpus derived from transcribed and translated video recordings

License: MDC DLA-1.0

Locale: haz, en

Task: MT

Format: TSV

Size: 434.04 KB

MDC DatasetsCompensated

🚧 Hazargi-English Speech Translation Corpus 🚧

A Hazargi speech translation dataset with approximately 10 hours of Hazargi audio with English text translations.

License: MDC DLA-1.0

Locale: haz, en

Task: MT

Format: MP3, TSV

Size: 511.58 MB

MDC DatasetsCompensated

🚧 Khowar Multimodal Dataset 🚧

A Khowar multimodal dataset with 10 hours of synchronized video, audio, transcriptions, speech segmentation, and ELAN annotations.

License: MDC DLA 1.0

Locale: khw, en

Task: NLP

Format: MP4, WAV, EAF

Size: 167.23 GB

The University of Melbourne

Hawrami Kurdish TTS dataset 1.0

This dataset contains high-quality single-speaker audio recordings in Hawrami Kurdish (Hewrami, ISO 639-3:hac), also known as the Gorani language, intended for building Text-to-Speech (TTS) and Automatic Speech recognition (ASR) systems. The dataset comprises 5 hours and 15 minutes of aligned audio and text data. Hawrami is classified as Definitely Endangered by UNESCO.

License: CC-BY-4.0

Locale: hac

Task: TTS

Format: WAV

Size: 706.11 MB

Keblagh e Azergi

Elkhani Hazargi Literature Corpus

Hazargi literary corpus (~0.5M tokens) of poetry, folklore, and prose texts representing Hazara linguistic and cultural heritage.

License: CC-BY-NC-4.0

Locale: haz

Task: NLP

Format: TXT

Size: 2.46 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

Common Voice

Common Voice Scripted Speech 26.0 - Hazargi

A collection of read speech recordings in Hazargi.

License: CC0-1.0

Locale: haz

Task: ASR

Format: MP3

Size: 175.19 MB

Tamazight NLP

Tamazight Open Speech Dataset

This dataset provides a parsed, formatted, and ready-to-use Amazigh Voice Dataset. It contains voice recordings and corresponding text transcripts in Standard Moroccan Amazigh (ⵜⴰⵎⴰⵣⵉⵖⵜ ⵜⴰⵏⴰⵡⴰⵢⵜ ⵜⴰⵎⵓⵔⴰⴽⵓⵛⵜ) intended for training Automatic Speech Recognition (ASR) and Text-to-Speech (TTS) models.

License: Apache-2.0

Locale: zgh

Task: ASR

Format: WAV, JSONL

Size: 459.09 MB

Keblagh e Azergi

Keblagh-e-Azergi Hazargi literature corpus

This corpus is a collection of more than one hundred thousand tokens of Hazargi language. The corpus contains work of literature, poems, folk and short stories and dramas. The data is being shared with the approval of the authors. It claims to support linguistic research, language technology development and cultural preservation.

License: CC-BY-NC-4.0

Locale: haz

Task: NLP

Format: TXT

Size: 193.28 KB

Institute of Finno-Ugric/Uralic Studies, University of Hamburg

INEL Dolgan Speech Corpus

A 13-hour supervised Speech-to-Text dataset for Dolgan, an endangered Turkic language. Features sentence-level audio aligned with text transcriptions.

License: CC-BY-NC-SA-4.0

Locale: dlg

Task: ASR

Format: TSV, MP3

Size: 583.34 MB

The University of Melbourne

Central Kurdish TTS dataset 1.0

This dataset contains high-quality single-speaker audio recordings in Central Kurdish (ckb), intended for building Text-to-Speech (TTS) and Automatic Speech recognition (ASR) systems. The dataset comprises 2 hours and 18 minutes of aligned audio and text data.

License: CC-BY-4.0

Locale: ckb

Task: TTS

Format: wav

Size: 293.45 MB

MirasAI

Multispeaker Hindi ASR Dataset

A Hindi audio dataset with corresponding transcriptions, designed for speech recognition and transcription tasks.

License: CC-BY-NC-SA-4.0

Locale: hin

Task: ASR

Format: OGG , SRT

Size: 63.03 MB

MDC DatasetsCompensated

🚧 Khowar Corpus for Speech Recognition 🚧

A Khowar ASR dataset with approximately 10 hours of audio with transcriptions.

License: MDC DLA-1.0

Locale: khw

Task: ASR

Format: MP3, TSV

Size: 245.00 MB

MDC Curators

Multilingual Audio Visual Speech (Lip-Reading) Dataset

Anonymised mouth-only video paired with audio and transcribed text in multiple languages.

License: CC-BY-SA-4.0

Locale: mul

Task: OTH

Format: WEBM, TSV

Size: 32.18 MB

Community

Awal Tamazight Dataset

A compilation of monolingual and parallel Tamazight language datasets created by CIEMEN as part of the Awal community language technology project.

License: CC-BY-4.0

Locale: zgh

Task: LM

Format: TSV, JSON, TXT

Size: 11.57 MB

Community

Urdu Multi-Speaker TTS Dataset

An Urdu multi-speaker TTS dataset distributed in 36 zip files, each containing audio files and a TSV mapping file, with approximately 10 hours of speech.

License: CC-BY-NC-4.0

Locale: urd

Task: TTS

Format: WEBM, TSV

Size: 514.54 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

Kaleem Art Press

Multilingual Religious Parallel Corpus (Kaleem Art Press)

This dataset is a multilingual parallel sentences corpus containing 6,465 aligned sentence units with approximately 0.98 million words, curated from Kaleem Art Press archives. It includes parallel religious text data in Arabic, Urdu, Saraiki (standard and dialectal), Punjabi (Shahmukhi), and English, supporting research in machine translation, comparative linguistics, digital humanities, and low-resource language studies.

License: CC-BY-SA-4.0

Locale: mul

Task: MT

Format: CSV

Size: 2.27 MB

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

MDC DatasetsCompensated

🚧 Multi-Subject Human Pose & Facial Expression Dataset (Studio, 4K) 🚧

A studio‑captured computer‑vision dataset of 16 human subjects

License: MDC DLA-1.0

Locale: zxx

Task: CV

Format: JPEG, CSV

Size: 68.50 GB

Institute of African Digital Humanities

Batanga-ALCAM-MultimodalDataset

Multimodal dataset for Batanga : lexical datasheet, audio recordings, and an audio alignment file for Banoho. Bapuku, forthcoming.

License: NOODL-1.0

Locale: bnm

Task: NLP

Format: MP3, TSV

Size: 18.70 MB

All Datasets

Not sure where to start?

🚧 Hazargi Multimodal Dataset 🚧

🚧 Hazargi Corpus for Speech Recognition 🚧

🚧 Northern Pashto Multimodal Dataset 🚧

🚧 Hazargi-English Parallel Text Corpus 🚧

🚧 Hazargi-English Speech Translation Corpus 🚧

🚧 Khowar Multimodal Dataset 🚧

Hawrami Kurdish TTS dataset 1.0

Elkhani Hazargi Literature Corpus

Bati-MultiDialectalASR-Dataset

Common Voice Scripted Speech 26.0 - Hazargi

Tamazight Open Speech Dataset

Keblagh-e-Azergi Hazargi literature corpus

INEL Dolgan Speech Corpus

Central Kurdish TTS dataset 1.0

Multispeaker Hindi ASR Dataset

🚧 Khowar Corpus for Speech Recognition 🚧

Multilingual Audio Visual Speech (Lip-Reading) Dataset

Awal Tamazight Dataset

Urdu Multi-Speaker TTS Dataset

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

Multilingual Religious Parallel Corpus (Kaleem Art Press)

RFE/RL Tatar-Bashkir News Text Corpus

🚧 Multi-Subject Human Pose & Facial Expression Dataset (Studio, 4K) 🚧

Batanga-ALCAM-MultimodalDataset