Speech recognition?

[pachpage]

Although getUserMedia is helpful for getting camera input, I'm developing a web app that uses speech recognition. I looked into Pocketsphinx.js, but it's dictionary seems small and when I tried running the demo in Bowser, it responded in the Status section of the page "No live audio input in this browser."

So, since getUserMedia won't cut it, is there any way to use webkitSpeechRecognition in Bowser?

[superdump]

That depends whether it is supported by UIWebView or WKWebView.

We have not implemented any WebAudio support so I suspect that is why speech recognition web apps may not work.

[stefanalund]

@pachpage we just landed WKWebView support. Please have another look.

[binki]

I think Bowser’s job is to just get you the audio stream so you can send it somewhere with WebRTC, e.g., a server which can handle speech recognition. iOS as of even 10.2 does not appear to have webkitSpeechRecognition or SpeechRecognition in Safari. So guessing this request is outside of the scope of Bowser and should be closed.