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

Filtered by

Search results for “covost 2”

100 Datasets

Mozilla Common Voice

Common Voice Spontaneous Speech 5.0 - Arvanitika

A collection of spontaneous responses to questions in Arvanitika.

License: CC0-1.0

Locale: aat

Task: ASR

Format: MP3

Size: 46.85 MB

THA_ATLAS

StackOverflow Vector Dataset - Cobol

Language-specific StackOverflow vector dataset for Cobol.

License: CC-BY-SA-4.0

Locale: en-US

Task: RAG

Format: Parquet

Size: 192.29 MB

Open Home Foundation

Gosia 1.0

Text to speech dataset for Polish, female speaker, approximately 2 hours of read speech.

License: CC0-1.0

Locale: pl-PL

Task: TTS

Format: WEBM

Size: 39.75 MB

Center za jezikovne vire in tehnologije Univerze v Ljubljani

SI-NLI

SI-NLI (Slovene Natural Language Inference Dataset) contains 5,937 human-created Slovene sentence pairs (premise and hypothesis) that are manually labeled with the labels "entailment", "contradiction", and "neutral". We created the dataset using sentences that appear in the Slovenian reference corpus ccKres (http://hdl.handle.net/11356/1034). Annotators were tasked to modify the hypothesis in a candidate pair in a way that reflects one of the labels. The dataset is balanced since the annotators created three modifications (entailment, contradiction, neutral) for each candidate sentence pair. The dataset is split into train, validation, and test sets, with sizes of 4,392, 547, and 998. We used Slovenian pre-trained language models to create splits, thereby ensuring that difficult and easy instances are evenly distributed in all three subsets. The dataset is released in a tabular TSV format. The README.txt file contains a description of the attributes. Only the hypothesis and premise are given in the test set (i.e. no annotations) since SI-NLI is integrated into the Slovene evaluation framework SloBENCH (https://slobench.cjvt.si/). If you use the dataset to train your models, please consider submitting the test set predictions to SloBENCH to get the evaluation score and see how it compares to others.

License: CC-BY-NC-SA-4.0

Locale: sl

Task: NLU

Format: TSV

Size: 392.44 KB

Mozilla Common Voice

Common Voice Scripted Speech 27.0 - Loja Highland Kichwa

A collection of read speech recordings in Loja Highland Kichwa.

License: CC0-1.0

Locale: qvj

Task: ASR

Format: MP3

Size: 221.75 MB

Mozilla Common Voice

Common Voice Scripted Speech 27.0 - Kotokoli

A collection of read speech recordings in Kotokoli.

License: CC0-1.0

Locale: kdh

Task: ASR

Format: MP3

Size: 173.72 MB

Weekly dataset digest

Receive a brief weekly digest of new datasets similar to the ones you’ve downloaded.

Yes, notify me

Mozilla Common Voice

Common Voice Scripted Speech 27.0 - Tshivenda

A collection of read speech recordings in Tshivenda.

License: CC0-1.0

Locale: ve

Task: ASR

Format: MP3

Size: 1.33 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

Mozilla Common Voice

Common Voice Scripted Speech 27.0 - Northern Sotho

A collection of read speech recordings in Northern Sotho.

License: CC0-1.0

Locale: nso

Task: ASR

Format: MP3

Size: 3.39 MB

Mozilla Common Voice

Common Voice Scripted Speech 27.0 - Parkari Koli

A collection of read speech recordings in Parkari Koli.

License: CC0-1.0

Locale: kvx

Task: ASR

Format: MP3

Size: 219.49 MB

Mozilla Common Voice

Common Voice Spontaneous Speech 5.0 - Toba Qom

A collection of spontaneous responses to questions in Toba Qom.

License: CC0-1.0

Locale: tob

Task: ASR

Format: MP3

Size: 173.39 MB

Mozilla Common Voice

Common Voice Scripted Speech 27.0 - Serbian

A collection of read speech recordings in Serbian.

License: CC0-1.0

Locale: sr

Task: ASR

Format: MP3

Size: 246.90 MB

MDC Curators

LibriVox Croatian TTS Male Voice

4 hours of sentence-aligned speech/text from "Priče iz Davnine" on LibriVox

License: CC0-1.0

Locale: hr

Task: TTS

Format: MP3, TXT, TSV

Size: 377.60 MB

Mozilla Common Voice

Common Voice Scripted Speech 27.0 - Romansh Vallader

A collection of read speech recordings in Romansh Vallader.

License: CC0-1.0

Locale: rm-vallader

Task: ASR

Format: MP3

Size: 113.42 MB

Taruen

Chuvash TTS

A ~5-hour speech dataset for Chuvash Text-to-Speech (TTS) research, featuring a single female speaker reading news and digits at a rapid tempo.

License: CC-BY-SA-4.0

Locale: cv

Task: TTS

Format: PARQUET

Size: 854.02 MB

Mozilla Common Voice

Common Voice Scripted Speech 27.0 - Nüpode Huitoto

A collection of read speech recordings in Nüpode Huitoto.

License: CC0-1.0

Locale: hux

Task: ASR

Format: MP3

Size: 230.34 MB

Mozilla Common Voice

Common Voice Scripted Speech 27.0 - Macedonian

A collection of read speech recordings in Macedonian.

License: CC0-1.0

Locale: mk

Task: ASR

Format: MP3

Size: 1.02 GB

Mozilla Common Voice

Common Voice Spontaneous Speech 5.0 - Wixárika

A collection of spontaneous responses to questions in Wixárika.

License: CC0-1.0

Locale: hch

Task: ASR

Format: MP3

Size: 199.53 MB

Mozilla Common Voice

Common Voice Scripted Speech 27.0 - Erzya

A collection of read speech recordings in Erzya.

License: CC0-1.0

Locale: myv

Task: ASR

Format: MP3

Size: 65.17 MB

Institute of African Digital Humanities

Bamun-French Parallel Corpus 2.0

This dataset is an extended and updated version of the "Bamun-French Parallel Corpus 1.1", a parallel corpus of 4,444 lines in Bamun and French.

License: NOODL-1.0

Locale: bax

Task: MT

Format: TSV

Size: 184.29 KB

Mozilla Common Voice

Common Voice Scripted Speech 27.0 - Indus Kohistani

A collection of read speech recordings in Indus Kohistani.

License: CC0-1.0

Locale: mvy

Task: ASR

Format: MP3

Size: 448.96 MB

Community

CV Korean Test 25.0 - Noise-Augmented (SCAI)

A noise-augmented version of the Mozilla Common Voice Korean test split for robust ASR evaluation under realistic acoustic conditions.

License: CC0-1.0

Locale: ko

Task: ASR

Format: MP3, JSONL

Size: 21.01 MB

Mozilla Common Voice

Common Voice Scripted Speech 27.0 - Tush

A collection of read speech recordings in Tush.

License: CC0-1.0

Locale: bbl

Task: ASR

Format: MP3

Size: 247.84 MB

RFE/RL

RFE/RL Chechen News Text Corpus

Longitudinal Chechen news corpus from Radio Marsho (2006-2026) with over 30,000 articles and 8.4M tokens.

License: CC-BY-NC-SA-4.0

Locale: ce

Task: NLP

Format: TXT

Size: 28.29 MB