JavaScript Speech Recognition

davidwalsh.name

14 points·hbhakhra··5 comments

> Chrome ends the listener after a given amount of time, so you'll need to hook into the end event to restart the speech listener

Which will burn up your battery on a mobile device. Chrome implements this by piping your audio stream to their servers. Not saying it should be avoided completely, just that you should know this going in.

Also, it is not enabled by default in Firefox, you have to enable it in about:config, which means you certainly can't count on it as a public user-facing feature. It's not implemented in any other browsers, as far as I'm aware.

[0] https://developer.mozilla.org/en-US/docs/Web/API/SpeechRecog...

There is this library as well: https://github.com/TalAter/annyang

He mentions it in the article. And it is really only an abstraction over the browser native SpeechRecognition API, it doesn't work in browsers that don't have it.

This was pretty mind blowing to me to see that speech recognition is built right into the browser now.

The last time I played with it was 2 years ago. I'd love to use it, but it's a resource hog in Chrome, Firefox doesn't enable it by default, and the other browsers don't implement it at all. Seems to be a bit of a chicken-or-egg problem with browser vendors not putting effort into over no apparent developer interest.

Reply on news.ycombinator.com