Common Voice is part of Mozilla's initiative to help teach machines how real people speak.
Metadata and versioning details for the Common Voice dataset
Script for bundling Common Voice (https://voice.mozilla.org) clips by language
Scraping Wikipedia for fair use sentences
Tool to collect and review sentences for Common Voice
Mozilla Voice Community Playbook
Tooling for producing French dataset for Common Voice
Automation for generating the common voice corpora
Different analysis and files from wikipedia text analysis
A living document outlining a methodological approach for building read speech sentence corpora.
All efforts around Mandarin dataset
Voicebot for contributing voice snippets to voice.mozilla.org
A Redux binding for React Router v4
This is where we organize the work around Common Voice project
This is a repo that will contain all the reviewed sentences collected by the global sprint.