Datasets | Mozilla Data Collective

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

Common Voice

Common Voice Scripted Speech 27.0 - Dhatki

A collection of read speech recordings in Dhatki.

License: CC0-1.0

Locale: mki

Task: ASR

Format: MP3

Size: 187.84 MB

MirasAI

Noakhalian (নোয়াখাইল্লা) Text Corpus

A 500K token corpus of Noakhalian linguistic data including drama scripts and cultural texts for professional NLP research and language preservation.

License: CC-BY-NC-4.0

Locale: oak

Task: NLP

Format: TXT, DOCX

Size: 3.61 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

All Datasets

Not sure where to start?

Common Voice Scripted Speech 27.0 - Kotokoli

Common Voice Scripted Speech 27.0 - Dhatki

Noakhalian (নোয়াখাইল্লা) Text Corpus

Gawri (گاؤری) Magazine Corpus