With speech recognition and response systems making big headways and the rising acceptance and usage of AI-driven assistants on phones and sensible speakers, the tech marketplace has realized that voice-based interaction is forming a massive element contributing towards the evolving future of human interaction with machines. However, building a voice recognition system for an app or gadget is not an easy task neither is it cheap because it requires a great deal of data to learn from.
Capitalising on this opportunity, the team at Mozilla, makers of Firefox browser, has come forward with an initiative, Common Voice, which is a project to crowdsource voice samples and validate them. It is seeking volunteers to contribute about 10,000 hours of audio in various accents to create its database to be freely available to developers who want to train speech-to-text systems for their projects. The database will be released later in 2017.
In order to contribute, one needs to head over to Common Voice site and record oneself reading a few sentences. One can also listen to others’ recordings and verify if they are accurate. iPhone users can make use of an iPhone app to record on the go. Privacy concerns of volunteers are taken care of and they do not need to be concerned about their personal information being revealed while contributing to the platform.
Do you think Mozilla's Common Voice initiative will be useful? Share your thoughts in the comments.
Rojenx is a leading concept artist who work appears in games and publications
Check out his personal gallery here