Mozilla launched the first open source voice recognition engine

Mozilla launched Common Voice in mid-June, this project is still in the training phase, users will help 'teach' the engine by reading short texts or verifying the accuracy of the recordings. before.

Mozilla wants your voice to teach the voice recognition engine

The goal of this is to collect at least 10,000 hours of voice recording, so Mozilla's engineers thought it was enough to teach the voice recognition system. Mozilla said it plans to launch Common Voice's database in open source by the end of 2017.

The reason for their participation in this project is that the open source voice recognition system is still lacking in the market. Currently, all of these engines are the company's own products, such as Amazon (Alexa), Apple (Siri) or Microsofot (Cortana), which are just a few names.

Mozilla launched the first open source voice recognition engine Picture 1
Common Voice is the first open source voice recognition engine project

Open source voice recognition engine will help small developers

'I think this will encourage creativity,' Mozilla said, 'so we created the Common Voice project, a project that brings voice recognition to people.'

Common Voice will bring voice recognition technology to everyone, including ordinary developers, who can't afford to spend millions of dollars on technologies like Cortana.

Those who wish to contribute their voice to the Common Voice database can participate at the project website. https://voice.mozilla.org/ You don't need a good microphone or soundproof room to read these sample texts. Mozilla engineers say they want to collect data from the real environment, so even if there is a little noise or a dialectic voice is okay. The readings of the text are all English text.