Datasets | Mozilla Data Collective

This corpus consists of research works, academic theses, and scholarly papers, comprising approximately 185,000 tokens. It covers a range of academic topics and formal registers, reflecting standardized writing practices and disciplinary conventions. The corpus is intended to support linguistic analysis, discourse studies, and computational research on academic and research-oriented text.

Brahui Research Work Corpus

License: CC-BY-NC-SA-4.0

Locale: brh

Task: NLP

Format: TXT

Size: 1.13 MB

A ~150,000-word Brahui corpus from the monthly magazine Talar, covering editorials, essays, fiction, poetry, and socio-cultural commentary.

Talar (تلار) Barahui Magazine Corpus

License: CC-BY-NC-SA-4.0

Locale: brh

Task: NLP

Format: TXT

Size: 317.22 KB

A cleaned literary corpus in Western Balochi (Rakhshani) including articles, research, translations, and creative literature. (~1.1 m tokens)

Western Balochi Literature Cropus

License: CC-BY-NC-4.0

Locale: bgn

Task: NLP

Format: TXT

Size: 2.26 MB

A ~1.02M-token Balochi newspaper corpus from NAWA-E-WATAN, representing contemporary journalistic and public discourse.

NAWA-E-WATAN Balochi Newspaper Corpus

License: CC-BY-NC-4.0

Locale: bgn

Task: NLP

Format: TXT

Size: 1.43 MB

Collaborative Action For Research & Development (CARD)

Gawri (گاؤری) Magazine Corpus

The Gawri (گاؤری) Magazine Corpus is a curated collection of monthly Gawri magazine text (~67,724 tokens) that supports research and language technology.

License: CC-BY-NC-4.0

Locale: gwc

Task: NLP

Format: TXT

Size: 146.71 KB

A scripted speech collection of audio recordings featuring the distinctive Ngapak dialect from the North Coast of Central Java (Pantura) Province, Indonesia.

TTS Javanese - Ngapak Dialect

License: CC-BY-SA-4.0

Locale: jav

Task: TTS

Format: WEBM, TSV

Size: 567.12 MB

Universitas Gadjah Mada

Jember Javanese Spontaneous Speech Corpus

A 10-hour spoken dataset of native Javanese speakers from Jember, East Java, representing the Jember dialect and Pandhalungan variety in natural speech.

License: CC-BY-NC-SA-4.0

Locale: jav

Task: ASR

Format: MP3, TSV

Size: 271.65 MB

Ok nemi totlahtool

Zacatlán Tepetzintla Nahuatl Transcriptions

The most up-to-date version of ongoing transcription effort corresponding to the Zacatlan Tepetzinla Nahuatl Audio dataset.

License: CC-BY-ND-4.0

Locale: nhi

Task: ASR

Format: TRS

Size: 320.28 KB

Ok nemi totlahtool

Zacatlán Tepetzintla Nahuatl Audio

Approximately 114 hours of recorded audio of Zacatlán-Ahuacatlán-Tepetzintla Nahuatl language (Glottocode zaca1241).

License: CC-BY-ND-4.0

Locale: nhi

Task: ASR

Format: WAV

Size: 50.19 GB

The dataset consists of 3 hours and 16 minutes of denoised audio clips, each paired with text and read by a single Bulu speaker.

Bulu-TTS-Dataset 1.0

License: NOODL-1.0

Locale: bum

Task: TTS

Format: MP3, TSV

Size: 87.40 MB

TTS dataset that uses everyday Sasak language in informal contexts with various topics.

TTS Sasak Language

License: CC-BY-SA-4.0

Locale: sas

Task: TTS

Format: WEBM, TSV

Size: 293.92 MB

Betawi TTS of Cultural Language (BEKAL) this dataset uses the Betawi dialect of West Java with Indonesian code-mixing and code-switching.

Betawi TTS of Cultural Language (BEKAL)

License: CC-BY-SA-4.0

Locale: bew

Task: TTS

Format: WEBM, TSV

Size: 309.99 MB

Digital Divide Data

Khmer ASR Cultural Dataset (V2)

106.53 hours manually curated speech-text pairs by native speakers in Khmer language about Cambodian cultural topics. On average, each recording is 8.42 seconds with the standard deviation of 3.39. Speaker metadata (gender, age group, and origin city) is provided. - Language: Khmer (khm). - Source(s): Native speakers from Cambodia (4 females, 4 males). The utterances were manually generated based on topics and subtopics listed in metadata. - Domain(s): Cultural domain, with a total of 2 topics and 118 subtopics. - Size: 45.57k data instances - WAV file names are formatted as: `{speaker_id}_khm_{sentence_id}.wav`.

License: CC-BY-SA-4.0

Locale: khm

Task: ASR

Format: WAV

Size: 35.86 GB

Taruen

Tatar Folklore Text Corpus

A 485k-word Tatar folklore corpus from 20th-century field recordings, selected from 5 academic volumes to prioritize contemporary linguistic usage.

License: CC0-1.0

Locale: tt

Task: NLP

Format: TXT

Size: 1.40 MB

This dataset comprises a compilation of cultural narratives and children’s stories from Southeast Sulawesi, Indonesia, presented in the Tolaki language.

TTS-Tolaki

License: CC-BY-NC-SA-4.0

Locale: lbw

Task: TTS

Format: WEBM, TSV

Size: 249.04 MB

This dataset is a compilation of spontaneous Mandar speech featuring Indonesian code-switching.

Mandar Spontaneous Speech

License: CC-BY-NC-4.0

Locale: mdr

Task: ASR

Format: MP3, TSV

Size: 534.45 MB

This dataset consists of audio recordings and textual data in Central Javanese (Semarang dialect) including Indonesian and English code-switching.

TTS Central Javanese

License: CC-BY-SA-4.0

Locale: jav

Task: TTS

Format: WEBM, TSV

Size: 440.11 MB

This dataset comprises audio recordings of scripted speech in Javanese of Lumajang Dialect from East Java of Indonesia.

TTS Javanese-Lumajang Dialect

License: CC-BY-SA-4.0

Locale: jav

Task: TTS

Format: WEBM, TSV

Size: 684.32 MB

This dataset is an updated version of the 'Adamawa Fulfulde–French Parallel Corpus of Narratives 1.1'.

Adamawa Fulfulde-French Parallel Corpus of Narratives 1.2

License: NOODL-1.0

Locale: fub

Task: MT

Format: TSV

Size: 112.17 KB

The dataset consists of four hours of high-quality audio clips, each paired with text and read by a single speaker.

Ewondo-TTS-Dataset

License: NOODL-1.0

Locale: ewo

Task: TTS

Format: MP3, TSV

Size: 152.70 MB

This dataset is an updated version of the "Bamun-French Parallel Corpus", a parallel corpus of texts in Bamun (Shupament) and French.

Bamun-French Parallel Corpus 1.1

License: NOODL-1.0

Locale: bax

Task: MT

Format: TSV

Size: 99.78 KB