Common Voice

In order to better support a globally distributed community of dataset users, developers, language activists and voice contributors the Common Voice team is launching a Discord server.

The Common Voice navbar will be changing to allow for a 2025 update that introduces Spontaneous Speech, a new mode of contribution and new dataset.

Op 4 novimber 2024 wiziget de manier wêrop meiwurkers oanmelde by Common Voice. Dizze wiziging hat gefolgen foar brûkers dy’t harren op dit stuit oanmelde mei Github.

The Common Voice data collection platform has added a way to submit multiple CC0 sentences at once into the text corpus.

Details oer ferzje 19.0 fan de iepen spraakgegevensset fan Common Voice

Om it wurk fan Common Voice better te ferbinen mei de taalmienskippen dy’t it betsjinnet, siket it programma frijwilligers dy’t ynteressearre binne om te helpen by it stypjen fan yndividuele talen.

To better meet the needs of lower resource languages, Common Voice is making changes to the amount of localisation new languages on the platform need to complete before seeing their language go live on Common Voice.

De meast resinte ferzje fan de Common Voice-spraakgegevensset biedt mear as 30.000 oeren oan spraakgegevens, fergees ûnder in CC0-lisinsje.

An overview of Common Voice's roadmap, goals and priorities for 2024 and beyond.

The Mozilla Common Voice dataset is expanding the length of clips held in its speech corpus to 15 seconds.

It's getting easier to create domain-specific datasets on Common Voice - from health to education to climate

We are adding more inclusive gender options for the Common Voice dataset