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

Not sure where to start?

Describe what you are building and the Data Assistant will find the right datasets for you.

Try the Data Assistant

Filtered by

Search results for “Hazargi Multimodal Dataset”

100 Datasets

Institute of African Digital Humanities

Afaan Oromoo-TTS-Dataset

1,737 read-speech recordings from a single Afaan Oromo speaker, with MP3 audio and TSV sentence mappings. Suitable for TTS and ASR.

License: NOODL-1.0

Locale: gax

Task: TTS

Format: MP3, TSV

Size: 258.48 MB

Institute of African Digital Humanities

Teke-Laali-TTS-Dataset

The dataset contains paired audio and text resources for Teke-Laali, a Bantu language spoken in the Congo. It consists of seven folders containing a total of 9,069 audio clips from raw audio recordings, with a total duration of 7:01:50.126 (HH:MM:SS.mmm). Additionally, there are seven audio/text mapping files containing a total of 9,069 lines. The dataset is suitable for TTS tasks.

License: NOODL-1.0

Locale: lli

Task: TTS

Format: WAV, TSV

Size: 635.61 MB

Institute of African Digital Humanities

Baka-ALCAM-MultimodalDataset

Multimodal dataset for Baka (bkc): lexical datasheet, audio recordings, and audio-sentence mapping files for Baka lexicon and grammar.

License: NOODL-1.0

Locale: bkc

Task: NLP

Format: WAV, TSV

Size: 63.15 MB

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

INEL Selkup Speech Corpus

A 1.6-hour supervised Speech-to-Text dataset for Selkup, an endangered Samoyedic language. Features sentence-level audio-text alignments.

License: CC-BY-NC-SA-4.0

Locale: sel

Task: ASR

Format: TSV, MP3

Size: 45.46 MB

Institute of African Digital Humanities

Basaa-ASR-Dataset

ASR/TTS dataset for Basaa (bas): 2,498 audio clips (01:54:46) in AGLC orthography, 25 sessions, including Basaa-ba-Yabassi and Babimbi varieties.

License: NOODL-1.0

Locale: bas

Task: ASR

Format: TSV, MP3

Size: 76.89 MB

Institute of African Digital Humanities

Tupuri-Bango_TTS-Dataset (female voice)

Tupuri-Bango (tui) scripted TTS dataset: 2,034 clips (3h 53m 42s), single female speaker, AGLC orthography. Female-voice complement to the male-voice dataset.

License: NOODL-1.0

Locale: tui

Task: TTS

Format: MP3, TSV

Size: 169.22 MB

MirasAI

Saraiki 10 Hours TTS Dataset

A 10-hour Saraiki text-to-speech dataset consisting of recorded speech and aligned transcripts, designed for speech synthesis research and development.

License: CC-BY-NC-SA-4.0

Locale: srk

Task: TTS

Format: WEBM, TSV

Size: 584.44 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

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

INEL Nganasan Speech Corpus

A 38.5-hour Speech-to-Text dataset for Nganasan, an endangered Samoyedic language. Features audio aligned with text transcriptions.

License: CC-BY-NC-SA-4.0

Locale: nio

Task: ASR

Format: TSV, MP3

Size: 1.29 GB

DDD-Cambodia

Khmer ASR Cultural Dataset (Version 3 - Part 4)

Khmer ASR Cultural Dataset is licensed under CC-BY-SA-4.0. Please attribute Digital Divide Data when using it.

License: CC-BY-SA-4.0

Locale: khm

Task: ASR

Format: WAV

Size: 25.05 GB

Community

Mandar Balanipa Speech Corpus

This Mandar language dataset is text and audio data paired for languange documentation and linguistic research in technology.

License: CC-BY-NC-4.0

Locale: mdr

Task: TTS

Format: WEBM, TSV

Size: 281.70 MB

MDC Curators

CoVoST 2 English - Mongolian

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

License: CC-BY-NC-4.0

Locale: en, mn

Task: MT

Format: MP3, TSV

Size: 12.56 GB

DDD-Cambodia

Khmer ASR Cultural Dataset (Version 3 - Part 5)

Khmer ASR Cultural Dataset is licensed under CC-BY-SA-4.0. Please attribute Digital Divide Data when using it.

License: CC-BY-SA-4.0

Locale: khm

Task: ASR

Format: WAV

Size: 33.14 GB

Habib University

Burushaski-English Speech Translation Corpus

A linguistically informed Burushaski speech - English text corpus containing 14,970 short audio recordings paired with corresponding English text translations.

License: CC-BY-NC-SA-4.0

Locale: bsk

Task: MT

Format: WAV, TXT

Size: 921.47 MB

DDD-Cambodia

Khmer ASR Cultural Dataset (Version 3 - Part 7)

Khmer ASR Cultural Dataset is licensed under CC-BY-SA-4.0. Please attribute Digital Divide Data when using it.

License: CC-BY-SA-4.0

Locale: khm

Task: ASR

Format: WAV

Size: 29.93 GB

Habib University

Burushaski-English Speech Translation Corpus

A linguistically informed Burushaski speech - English text corpus containing 14,970 short audio recordings paired with corresponding English text translations.

License: CC-BY-NC-SA-4.0

Locale: bsk

Task: MT

Format: WAV, TXT

Size: 921.47 MB

MDC Curators

CoVoST 2 Mongolian - English

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

License: CC-BY-NC-4.0

Locale: mn, en

Task: MT

Format: MP3, TSV

Size: 225.93 MB

Field Matters

AIRI AI4TALK Evenki ASR

Evenki ASR data from AIRI AI4TALK with MP3 speech segments and IPA-like transcriptions in CSV.

License: CC-BY-SA-4.0

Locale: evn

Task: ASR

Format: CSV, MP3

Size: 572.87 MB

DDD-Cambodia

Khmer ASR Cultural Dataset (Version 3 - Part 6)

Khmer ASR Cultural Dataset is licensed under CC-BY-SA-4.0. Please attribute Digital Divide Data when using it.

License: CC-BY-SA-4.0

Locale: khm

Task: ASR

Format: WAV

Size: 17.57 GB

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

INEL Enets Speech Corpus

A 3.7-hour supervised Speech-to-Text dataset for Enets, an endangered Samoyedic language. Features sentence-level audio aligned with text transcriptions.

License: CC-BY-NC-SA-4.0

Locale: enf, enh

Task: ASR

Format: TSV, MP3

Size: 140.56 MB

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

INEL Kamas Speech Corpus

A 14-hour Speech-to-Text dataset for Kamas, an extinct Samoyedic language waiting to be revitalized. Features audio aligned with text transcriptions.

License: CC-BY-NC-SA-4.0

Locale: xas

Task: ASR

Format: TSV, MP3

Size: 376.64 MB

DDD-Cambodia

Khmer ASR Cultural Dataset (Version 3 - Part 3)

Khmer ASR Cultural Dataset is licensed under CC-BY-SA-4.0. Please attribute Digital Divide Data when using it.

License: CC-BY-SA-4.0

Locale: khm

Task: ASR

Format: WAV

Size: 30.80 GB

DDD-Cambodia

Khmer ASR Cultural Dataset (Version 3 - Part 2)

Khmer ASR Cultural Dataset is licensed under CC-BY-SA-4.0. Please attribute Digital Divide Data when using it.

License: CC-BY-SA-4.0

Locale: khm

Task: ASR

Format: WAV

Size: 34.07 GB

Institute of African Digital Humanities

Bulu-ASR-Dataset

Bulu (bum) scripted speech dataset: 819 clips from 8 speakers (predominantly female) across 9 recording sessions.

License: NOODL-1.0

Locale: bum

Task: ASR

Format: MP3, TSV

Size: 42.11 MB