Datasets | Mozilla Data Collective

Language-specific StackOverflow vector dataset for Erlang.

StackOverflow Vector Dataset - Erlang

License: CC-BY-SA-4.0

Locale: en-US

Task: RAG

Format: Parquet

Size: 588.72 MB

Language-specific StackOverflow vector dataset for Abap.

StackOverflow Vector Dataset - Abap

License: CC-BY-SA-4.0

Locale: en-US

Task: RAG

Format: Parquet

Size: 554.44 MB

Language-specific StackOverflow vector dataset for Shell.

StackOverflow Vector Dataset - Shell

License: CC-BY-SA-4.0

Locale: en-US

Task: RAG

Format: Parquet

Size: 2.58 GB

Language-specific StackOverflow vector dataset for SQL.

StackOverflow Vector Dataset - SQL

License: CC-BY-SA-4.0

Locale: en-US

Task: RAG

Format: Parquet

Size: 15.84 GB

Language-specific StackOverflow vector dataset for Swift.

StackOverflow Vector Dataset - Swift

License: CC-BY-SA-4.0

Locale: en-US

Task: RAG

Format: Parquet

Size: 9.30 GB

Language-specific StackOverflow vector dataset for Typescript.

StackOverflow Vector Dataset - Typescript

License: CC-BY-SA-4.0

Locale: en-US

Task: RAG

Format: Parquet

Size: 7.31 GB

Language-specific StackOverflow vector dataset for Vb.Net.

StackOverflow Vector Dataset - Vb.Net

License: CC-BY-SA-4.0

Locale: en-US

Task: RAG

Format: Parquet

Size: 4.36 GB

Language-specific StackOverflow vector dataset for Elixir.

StackOverflow Vector Dataset - Elixir

License: CC-BY-SA-4.0

Locale: en-US

Task: RAG

Format: Parquet

Size: 572.33 MB

Biomedical Urdu Named Entity Recognition dataset for advancing healthcare and clinical NLP research in Urdu.

Biomedical Urdu Named Entity Recognition (BioUNER)

License: CC-BY-NC-4.0

Locale: urd

Task: NLP

Format: CSV

Size: 336.69 KB

A large-scale Sindhi Part-of-Speech (SiPOS) tagging dataset for developing and evaluating NLP models in a low-resource language.

Sindhi Parts Of Speech (SiPOS)

License: CC-BY-NC-4.0

Locale: snd

Task: NLP

Format: TXT

Size: 874.04 KB

A Sindhi figurative language dataset for identifying and classifying idioms, metaphors, and proverbs, similies in a low-resource language.

Sindhi Figurative Language Dataset (SiNFluD)

License: CC-BY-NC-4.0

Locale: snd

Task: NLP

Format: JSON

Size: 90.65 KB

A benchmark Sindhi question answering dataset for advancing machine reading comprehension and NLP research in a low-resource language.

Sindhi Question-Answering Dataset (SdQuAD)

License: CC-BY-NC-4.0

Locale: snd

Task: LLM

Format: JSON

Size: 500.50 KB

A Sindhi named entity recognition dataset for building NLP tools in an underrepresented South Asian language.

Sindhi Named Entity Recognition-SiNER Dataset

License: CC-BY-SA-4.0

Locale: snd

Task: NLP

Format: TXT

Size: 2.51 MB

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

Ghomala (bbj) scripted TTS/ASR dataset: 997 clips (≈59m 24s) across 10 sessions, transcribed in AGLC orthography. Bandjoun, West Region, Cameroon.

Sample Ghomala-TTS-Dataset

License: NOODL-1.0

Locale: bbj

Task: TTS

Format: MP3, TSV

Size: 80.67 MB

3 WAV recordings (≈57 min) of traditional Rwandan panegyric poetry (ibisigo). EMAC project, CERDOTOLA, digitised 2015.

EMAC-Dataset-Rwanda

License: NOODL-1.0

Locale: RW

Task: OTH

Format: WAV, XML

Size: 339.53 MB

5 WAV recordings (≈50 min) of traditional vocal music from Gabon: song-fables and pastoral epic poetry. EMAC project, digitised 2015.

EMAC-Dataset-Gabon

License: NOODL-1.0

Locale: GA

Task: OTH

Format: WAV, XML

Size: 301.57 MB

Ngomba (jgo) scripted TTS dataset: 1033 clips (≈1h06m) from 11 sessions, transcribed in AGLC orthography. Bamboutos Division, West Region, Cameroon.

Sample-Ngomba-TTS-Dataset

License: NOODL-1.0

Locale: jgo

Task: TTS

Format: MP3, TSV

Size: 47.47 MB

5 WAV recordings (≈1h25m) of traditional vocal music from DRC: lullabies, song-fables, and epic music. EMAC project, digitised 2015.

EMAC-Dataset-DRC

License: NOODL-1.0

Locale: CD

Task: OTH

Format: WAV, XML

Size: 500.98 MB

2 WAV recordings (≈45 min) of traditional vocal music from the Republic of Congo: a lullaby and a song-fable. EMAC project, digitised 2015.

EMAC-Dataset-Congo

License: NOODL-1.0

Locale: CG

Task: OTH

Format: WAV, XML

Size: 260.86 MB

5 WAV recordings (≈57 min) of traditional vocal music from the Central African Republic: song-fables and epic music. EMAC project, digitised 2015.

EMAC-Dataset-CAR

License: NOODL-1.0

Locale: CF

Task: OTH

Format: WAV, XML

Size: 366.79 MB

32 WAV recordings (≈5h38m) of Cameroonian music: epics, Mvet, Bikutsi, lullabies, and song-fables in Basaa and Beti. EMAC project, digitised 2015.

EMAC-Dataset-Cameroon

License: NOODL-1.0

Locale: CM

Task: OTH

Format: WAV, XML

Size: 1.95 GB