Welcome to Mozilla Voice
The new home for machine-learning based voice technologies -- including software, tools, and datasets.
Common Voice is Mozilla's global crowdsourcing initiative to build multilingual open voice datasets that help teach machines how real people speak.
Mozilla Voice STT
Our open source Speech-to-Text engine, that can run on anything from an off-line Raspberry Pi 4 to a server class machine, implemented with Google’s TensorFlow framework.
Mozilla Voice TTS
Built with advanced machine learning methods (Tacotron), our open source Text-to-Speech engine enables high quality, natural and human-like voice synthesis.
Mozilla Voice Developer Tools provide several key components of a full open source voice "stack". These components can be readily integrated with existing applications and platforms as well as suitable complementary technologies from other sources.
Mozilla Voice is built developer-first; giving you the APIs, documentation and tools to provide you with more choice in voice.
Our intention is that, over time, Mozilla Voice will become a feature rich voice technology platform.
Mozilla Voice aims to contribute to a more diverse, decentralized, and innovative voice technology ecosystem through quality data, competitive, scalable, easy to use developer tools and inspiring products. In a world where voice experiences become the default way to interact with the internet and the technologies around us,
we want these voice experiences to be privacy-preserving, open, and accessible to everyone. Because competition and openness are healthy for innovation. Because under - resourced languages are an issue of access and equity. Because privacy and control matter. Especially over your voice.