Connect
Sign up to receive our monthly newsletter.
Inquiries
Have a question? Contact us here.
Job Opportunities
Learn about job openings.
Digital Public Goods Alliance - 2026
Sign up to receive our monthly newsletter.
Have a question? Contact us here.
Learn about job openings.
DPGs for AI Collection
All DPGs included in the DPGs for AI Collection must meet the requirements outlined in the DPG4AI collection criteria.
The information below reflects the self-reported responses provided as part of the assessment process.
Solution Type: Open Data
DPG Compliance & Profile Page: https://www.digitalpublicgoods.net/r/mozilla-common-voice-dataset
Description: A multilingual open voice dataset created by Mozilla Common Voice, an initiative to help teach machines how real people speak
Assessment Status: Under Review
Review Date: 2026-06-14
Category Fit:
Open Data (CC0-1.0). Mozilla Common Voice is a multilingual speech dataset explicitly created to train and evaluate speech technology (ASR/STT, language ID). Lifecycle roles: Data Acquisition & Preparation and Model Building & Training. https://commonvoice.mozilla.org/
AI Lifecycle Utility:
Documented Relevance/ Impact:
Designed-for-AI by mandate ('teach machines how real people speak'). One of the largest public multilingual voice corpora (130+ languages across releases) with demographic metadata (age, sex, accent). Standard training data for Whisper-class and community ASR models. Versioned corpora (…Corpus 22.0) on the site and HF (mozilla-foundation/common_voice_\*). Datasets: https://commonvoice.mozilla.org/en/datasets
Adoption Readiness Level:
L5 Productized / Plug-and-Play
Adoption Readiness Evidence:
Regular numbered releases (Corpus 2.0 → 22.0) with versioning/changelog metadata (https://github.com/common-voice/cv-dataset), per-language splits with TSV metadata, and plug-and-play HF dataset loaders. https://huggingface.co/datasets/mozilla-foundation/common_voice_17\_0
Interoperability Level:
L5 Croissant + API — Croissant metadata standard + programmatic API access
Interoperability Evidence:
MP3 audio + TSV metadata; distributed via Hugging Face datasets (mozilla-foundation/common_voice_\*) which exposes auto-generated Croissant metadata and the datasets API/streaming loader. ISO 639 language codes. https://huggingface.co/datasets/mozilla-foundation/common_voice_17\_0
Responsible Practices Level:
Equity & Inclusion:
Responsible Practices:
Inclusion & Autonomy:
Standards & Vocabularies:
L3 Full Disclosure — bias + representation + de-identification method + provenance documented (Croissant-RAI, Datasheet, or Data Statement)
Responsible Practices Evidence:
Demographic representation metadata (age/sex/accent) ships with the corpus, and known gender/language imbalances are documented. Collection context, consent, age limits, right-to-be-forgotten and de-identification (no PII in released audio/metadata) are documented in terms/privacy. Original data statement: Ardila et al., 'Common Voice: A Massively-Multilingual Speech Corpus' (LREC 2020). Terms: https://commonvoice.mozilla.org/en/terms
