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

University of Bamberg

Less is More Corpus

This the dataset associated with the paper "Less is More? The Role of Demographic Author Information in Emotion Classification of Ambiguous Text"

License: CC-BY-4.0

Locale: eng

Task: NLP

Format: CSV

Size: 100.24 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

MDC Community Concierge

VoxForge - Albanian

30 minutes (330 utterances) of read speech of Shqip (Albanian).

License: GPL-3.0

Locale: sq

Task: ASR

Format: WAV, TSV

Size: 113.74 MB

MDC Community Concierge

VoxForge - Ukrainian

1 hour (390 utterances) of read speech of Українська (Ukrainian).

License: GPL-3.0

Locale: uk

Task: ASR

Format: WAV, TSV

Size: 252.64 MB

MDC Community Concierge

VoxForge - Russian

15.5 hours (6412 utterances) of read speech of Русский (Russian).

License: GPL-3.0

Locale: ru

Task: ASR

Format: WAV, TSV

Size: 3.53 GB

MDC Community Concierge

VoxForge - Hebrew

32 minutes (307 utterances) of read speech of ‭עברית‬ (Hebrew).

License: GPL-3.0

Locale: he

Task: ASR

Format: WAV, TSV

Size: 126.98 MB

MDC Community Concierge

VoxForge - Persian

16 minutes (157 utterances) of read speech of ‭فارسی‬ (Persian).

License: GPL-3.0

Locale: fa

Task: ASR

Format: WAV, TSV

Size: 58.20 MB

MDC Community Concierge

VoxForge - French

39 hours (23752 utterances) of read speech of Français (French).

License: GPL-3.0

Locale: fr

Task: ASR

Format: WAV, TSV

Size: 9.31 GB

MDC Community Concierge

VoxForge - Croatian

14 minutes (127 utterances) of read speech of Hrvatski (Croatian).

License: GPL-3.0

Locale: hr

Task: ASR

Format: WAV, TSV

Size: 59.27 MB

MDC Community Concierge

VoxForge - Portuguese

4.5 hours (4571 utterances) of read speech of Português (Portuguese).

License: GPL-3.0

Locale: pt

Task: ASR

Format: WAV, TSV

Size: 1.06 GB

MDC Community Concierge

VoxForge - Dutch

10.5 hours (8440 utterances) of read speech of Nederlands (Dutch).

License: GPL-3.0

Locale: nl

Task: ASR

Format: WAV, TSV

Size: 2.25 GB

MDC Community Concierge

VoxForge - Italian

20 hours (10643 utterances) of read speech of Italiano (Italian).

License: GPL-3.0

Locale: it

Task: ASR

Format: WAV, TSV

Size: 4.94 GB

MDC Community Concierge

VoxForge - Spanish

53 hours (23537 utterances) of read speech of Español (Spanish).

License: GPL-3.0

Locale: es

Task: ASR

Format: WAV, TSV

Size: 12.70 GB

Institute of African Digital Humanities

Efik-TTS-Dataset

Segmented Efik speech clips paired with text for TTS. Standard Calabar dialect, Cross River State, Nigeria. 1,362 clips, 5h 32m.

License: NOODL-1.0

Locale: efi

Task: TTS

Format: MP3, TSV

Size: 297.28 MB

Institute of African Digital Humanities

Igbo-TTS-Dataset

Segmented Igbo speech audio clips paired with text for Text-to-Speech (TTS), representing the standard spoken variety of southeastern Nigeria.

License: NOODL-1.0

Locale: ibo

Task: TTS

Format: MP3, TSV

Size: 172.77 MB

Institute of African Digital Humanities

Adamawa Fulfulde-TTS-Dataset

1,303 read-speech recordings from a single Adamawa Fulfulde speaker, with MP3 audio and a TSV sentence mapping. Suitable for TTS and ASR.

License: NOODL-1.0

Locale: fub

Task: TTS

Format: MP3, TSV

Size: 169.27 MB

MDC Curators

CoVoST 2 Catalan - English

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

License: CC-BY-NC-4.0

Locale: ca, en

Task: MT

Format: MP3, TSV

Size: 4.50 GB

MDC Curators

CoVoST 2 Welsh - English

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

License: CC-BY-NC-4.0

Locale: cy, en

Task: MT

Format: MP3, TSV

Size: 95.16 MB

MDC Curators

CoVoST 2 Arabic-English

The dataset includes 5736 audio files in Arabic and their translations in English.

License: CC-BY-NC-4.0

Locale: ar, en

Task: MT

Format: MP3, TSV

Size: 148.21 MB

MDC Community Concierge

VoxForge - German

32 hours (24685 utterances) of read speech of Deutsch (German).

License: GPL-3.0

Locale: de

Task: ASR

Format: WAV, TSV

Size: 11.59 GB

MDC Community Concierge

VoxForge - Greek

4 hours (1377 utterances) of read speech of Ελληνικά (Greek).

License: GPL-3.0

Locale: el

Task: ASR

Format: WAV, TSV

Size: 954.26 MB

MDC Community Concierge

VoxForge - Catalan

39 minutes of Catalan read speech.

License: GPL-3.0

Locale: ca

Task: ASR

Format: WAV, TSV

Size: 143.71 MB

0DIN by Mozilla

Public GenAI Vulnerability Disclosures

Weekly export of 0DIN's public GenAI vulnerability disclosures: jailbreaks, prompt injection, content manipulation, and other model-security findings.

License: CC-BY-4.0

Locale: en

Task: N/A

Format: JSONL

Size: 116.26 KB

MDC Community Concierge

VoxForge - Bulgarian

1 hour of read speech in Bulgarian, collected via the VoxForge project

License: GPL-3.0

Locale: bg

Task: ASR

Format: WAV, TSV

Size: 234.06 MB