New Speech Recognition Milestone

Image for post
Image for post

Microsoft announced it reached a new WER of 5.9% on an industry standard test. This is a 10%+improvement over it’s last record. However, Google has claimed (albeit through its own standards) a 4.9% WER.

It seems we’ll be seeing these improvements announced steadily over the coming year, as we surpass human WER parity and then start to get better at particular domains. Layering on top of this noise rejection, far field performance, and augmentation through training, we’re within two years of being understood better by a machine than by each other.

When that happens, we may start to rely on tools to analyze voice for us to give us feedback on things like truthfulness, emotion, or other measures to enhance our understanding of what people are telling us.

Independent daily thoughts on all things future, voice technologies and AI. More at

Get the Medium app

A button that says 'Download on the App Store', and if clicked it will lead you to the iOS App store
A button that says 'Get it on, Google Play', and if clicked it will lead you to the Google Play store