Web Speech API (input)
Enables web developers to incorporate speech recognition into their web pages.
Web Speech is on the backlog with a priority of Medium. You can track status changes at https://dev.windows.com/en-us/microsoft-edge/platform/status/webspeechapiinput
We are currently investigating Web Speech along with script-based accessibility as part of our longer term accessibility investments. This joins our wider ranging accessibility investments detailed in our Microsoft Edge for Developers blog post: https://blogs.windows.com/msedgedev/2016/02/03/2016-platform-priorities/
Robin Goupil commented
> Anonymous commented · July 07, 2015 04:20
> We have an example of the Speech Synthesis API here:
> In supported browsers that we have tested it does not show a security warning to users.
The example you posted concern "Speech Synthesis" (text to voice), not "Speech Recognition" (voice to text).
Speech Synthesis is as dangerous to the user as a video or audio in a web page, why would browser bother showing a warning to the user?
I am glad to see this feature is being developed, Edge might finally become a decent competitor :)
Please consider raising the priority for this API.
If Microsoft will get behind it, then it will build pressure on the other browser makers to support it, too.
There are very few browsers that support this API today, per http://caniuse.com/#feat=speech-recognition .
So long as Edge doesn't have this feature, I will be recommending to my users that they go and use Chrome for a better user experience. This is a shame as I felt Edge might actually give Chrome a run for it's money. I guess I was wrong.
We have an example of the Speech Synthesis API here:
In supported browsers that we have tested it does not show a security warning to users.
Kevin Lozandier commented
Accidently flagged this as inappropriate while trying to vote; sorry about that. This is needed & long overdue
We need this feature in Microsoft Edge
Jakub Jedryszek commented
According to spec it does: "User agents must only start speech input sessions with explicit, informed user consent."
Szymon Nowak commented
With "decades of research in speech recognition" (http://blogs.skype.com/2014/12/15/skype-translator-how-it-works) and upcoming Object RTC support, it shouldn't be that hard to add it to IE, right? :) It would be awesome if it was also possible to pass any media stream as input to SpeechRecognition constructor.
does it show security warnings to user?