We can't find the internet
Attempting to reconnect
Something went wrong!
Hang in there while we get back on track
Generate a DeepSpeech model with the help of your community How to get fun with teamwork
Learn how the Mozilla Italia community collaborated to generate an open-source DeepSpeech model for Italian, overcoming challenges and achieving success through community involvement and Mozilla employee support.
- The Mozilla Italia community joined the Common Voice project, aiming to generate a DeepSpeech model for Italian.
- The team worked on data collection, marketing, and documentation, with a focus on Italian language and culture.
- The project included testing and promoting the model through social media, events, and online platforms.
- The team also worked with Mozilla employees to improve the project and gained feedback from the community.
- The model is open-source and publicly available, with the goal of encouraging other communities to participate and create their own languages.
- The project faced challenges, such as ensuring diversity in recording data, managing accents and dialects, and verifying the quality of the model.
- The team used Bash scripts and Docker images to generate the model, and created an Android app for testing and usage.
- The project’s success was largely due to community involvement and the help of Mozilla employees, with special thanks to Alexander, who provided support and guidance.
- The team plans to continue improving the model and expanding to other languages, with a focus on promoting inclusivity and accessibility.