Mozilla Common Voice: “We are building an open and publicly available dataset of voices that everyone can use to train speech-enabled applications. We believe that large and publicly available voice datasets foster innovation and healthy commercial competition in machine-learning based speech technology. This is a global effort and we invite everyone to participate. Our aim is to help speech technology be more inclusive, reflecting the diversity of voices from around the world.”