Datasets | Mozilla Data Collective

Spanish word-frequency lists of wordfreq v3.1.1 (CC BY-SA 4.0) as TSV, plus the 16,740-word lexicon derived from them; used only for post-processing.

wordfreq Spanish word-frequency list — as used in Lost in Transcription (Spanish–Nahuatl)

License: CC-BY-SA-4.0

Locale: es

Task: NLP

Format: TSV, Markdown

Size: 1.87 MB

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

CoVoST 2 English - Slovenian

License: CC-BY-NC-4.0

Locale: en, sl

Task: MT

Format: MP3, TSV

Size: 12.56 GB

Institute of African Digital Humanities

Eton-TTS-Dataset

Eton (eto) single-speaker TTS dataset: 1,802 clips (~4h15m) by one female speaker, AGLC orthography with full tone marking.

License: NOODL-1.0

Locale: eto

Task: TTS

Format: WAV, TSV

Size: 1.04 GB

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

CoVoST 2 English - Tamil

License: CC-BY-NC-4.0

Locale: en, ta

Task: MT

Format: MP3, TSV

Size: 12.56 GB

A collection of spontaneous responses to questions in Gheg Albanian.

Common Voice Spontaneous Speech 5.0 - Gheg Albanian

License: CC0-1.0

Locale: aln

Task: ASR

Format: MP3

Size: 202.32 MB

The dataset includes audio files in English and their translations in Chinese (China).

CoVoST 2 English - Chinese (China)

License: CC-BY-NC-4.0

Locale: en, zh-CN

Task: MT

Format: MP3, TSV

Size: 12.56 GB

Weekly dataset digest

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

Yes, notify me

A collection of spontaneous responses to questions in Chinese (China).

Common Voice Spontaneous Speech 5.0 - Chinese (China)

License: CC0-1.0

Locale: zh-CN

Task: ASR

Format: MP3

Size: 121.91 KB

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

CoVoST 2 English - Turkish

License: CC-BY-NC-4.0

Locale: en, tr

Task: MT

Format: MP3, TSV

Size: 12.56 GB

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

CoVoST 2 English - Persian

License: CC-BY-NC-4.0

Locale: en, fa

Task: MT

Format: MP3, TSV

Size: 12.56 GB

Institute of African Digital Humanities

Ewondo_Fong_ALCAM-MultimodalDataset

Curated Ewondo (Fong) sentences with French translations, audio recordings, and sentence–audio alignments.

License: NOODL-1.0

Locale: ewo

Task: NLP

Format: MP3, TSV

Size: 16.80 MB

Collaborative Action For Research & Development (CARD)

IBT Torwali Wordlist

The IBT Torwali Wordlist contains approximately 20,000 unique entries in Torwali (ISO 639-3: trw), an under-documented Indo-Aryan language spoken in northern Pakistan. The dataset comprises standardized lexical entries covering core vocabulary, function words, and culturally salient terms, with consistent orthography and normalization suitable for linguistic and computational use. Entries are aligned with English and Urdu glosses, and include part-of-speech tag.

License: CC-BY-SA-4.0

Locale: trw

Task: NLP

Format: CSV

Size: 312.87 KB

The dataset covers a parallel corpus of 5,173 lines of literal translation text in Indonesian, Javanese multi-dialect, English, and Arabic.

Multidialect of JAV-IND-ENG-ARB Parallel Corpus

License: CC-BY-NC-SA-4.0

Locale: ind,jav,eng,arb

Task: MT

Format: TSV

Size: 566.75 KB

The languages represented in the corpus include Manggarain in the Middle Manggaraian dialect, Indonesian, English, and Arabic.

Middle Manggaraian Dialect of Manggarai-IND-ENG-ARB Parallel Corpus

License: CC-BY-SA-4.0

Locale: ind,mqy,eng,arb

Task: MT

Format: TSV

Size: 349.22 KB

A collection of spontaneous responses to questions in Tashlhiyt.

Common Voice Spontaneous Speech 5.0 - Tashlhiyt

License: CC0-1.0

Locale: shi

Task: ASR

Format: MP3

Size: 55.20 MB

Open Home Foundation

Joe 1.0

Text to speech dataset for English, male speaker, approximately 1 hour of read speech.

License: CC0-1.0

Locale: en-US

Task: TTS

Format: WEBM

Size: 75.78 MB

Open Home Foundation

Anna 1.0

Text to speech dataset for Hungarian, female speaker, approximately 1.5 hours of read speech.

License: CC0-1.0

Locale: hu-HU

Task: TTS

Format: WEBM

Size: 95.27 MB

Content X LabsCompensated

🚧 Thai Episodic Television // Rights-Cleared · 100 Hours 🚧

100 hours of rights-cleared Thai-language episodic television content (5–6 episodes/series), with English subtitles and structured metadata

License: CXL-DLA-1.0

Locale: TH

Task: N/A

Format: MP4, SRT, CSV

Size: 630.79 GB

A collection of read speech recordings in Chinese (China).

Common Voice Scripted Speech 27.0 - Chinese (China)

License: CC0-1.0

Locale: zh-CN

Task: ASR

Format: MP3

Size: 21.40 GB

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

MirasAI

Wakhi Literature Corpus

A literature corpus of Wakhi language texts, preserving the language’s literary, cultural, and linguistic heritage for research and low-resource NLP.

License: CC-BY-ND-4.0

Locale: wbl

Task: NLP

Format: DOCX, TXT

Size: 74.83 KB

This dataset consists of 1,000 UK bank sort codes read out aloud by a single male speaker of UK English from the Midlands.

UK Sort Codes - ASR Evaluation

License: CC-BY-4.0

Locale: en-GB

Task: ASR

Format: WEBM, TSV

Size: 23.76 MB

Chana

Awajún: Phonetic Word List

150 Awajún lexical items with WAV recordings and Praat TextGrids, part of a controlled comparative phonetic dataset of three Peruvian languages.

License: Chana 2.0

Locale: agr

Task: N/A

Format: WAV, TextGrid, MD, CSV

Size: 70.83 MB

MirasAI

Associated Press Pakistan 21.4 Million Pashto Corpus

Pashto news corpus from the Associated Press of Pakistan, containing 99,495 articles and 21.4 million words from 2015–2026.

License: CC-BY-NC-SA-4.0

Locale: pus

Task: NLP

Format: TXT

Size: 38.81 MB