Speech to text in html text areas, or in TinyMCE

I'm looking for ideas how to do this:  Speak into your PCs microphone, and text appears in the input field, or preferably within the TinyMCE editor.

The only thing I've found on this so far, is a speech to text API that the folks at Google were working on a few years ago that worked in the Chrome browser.  Like so many Google projects, they seem to have abandoned it.  I was just wondering if anyone here has ever given this subject any thought, or have any experience with it.

The scenario is this:

I have a vBulletin site that I am going to migrate to Dolphin 7.2.   One of the long standing members of my vBulletin site, is an older woman that over the years, has contributed a lot of good content to the site.  She also enjoyed keeping in touch with all her friends on a MLB message board.  She's never been an online text chat/video chat person.  She makes and replies to forum posts when she has the time, so no type of live communication with her friends would be suitable.  Forums are the best form of communication for her, and many of her friends.

Here's the problem: Over the past several years, for various reasons, it has become very difficult and often painful for her to type messages out on a keyboard.  What I have typed so far, might take her 45 painful minutes.  Since she has contributed so much content to my vBulletin site, I figure I owe her one, and would like to come up with something to make it much easier to enter a forum post, and I'd like it to work within Dolphin.

If the speech to text idea is not viable, the fallback idea would be to add audio/video post capability to forum topics.  We already have video comments, so this might be the most easily implemented solution.  I strongly prefer the speech to text solution due to the nature of a lot of content that needs to be posted.

Any insight would be appreciated.

Hey HL, speech to text does work in 7.1.x (at least from Android)
I haven't tried it on earlier version.  Try to find out what device she is using, that would be a good starting point.

Also, the fallback idea is the EXACT same thing I've been trying to figure out since last year.
 

Noble idea. Maybe the following can help you in the right direction...

CMU Sphinx (active development)

http://cmusphinx.sourceforge.net/

More information about how to connect to CMU Sphinx using PHP...

http://stackoverflow.com/questions/25666760/connecting-to-cmu-sphinx-using-php

A client side solution may work the best.  I don't know of any open source speech to text.  Dragon Naturally Speaking has been on the market for many years; so active development has improved the speech to text translation.  Like any such project one has to go through a training process so the software can learn the speech patters of the user, it presents text that the user repeats so the software learns the user's voice patterns.

Check out this page using Chrome.

https://www.google.com/intl/en/chrome/demos/speech.html

It works reasonably well.  I haven't  dug into how it works though.

 I did make mention of a keyboard. By inference, that means desktop PC.  Anyone with arthritic or other conditions that make typing difficult, will not likely enjoy entering forum posts, or otherwise interacting with a website on a tiny cell phone screen.    We're talking old folks here... that means big keyboards, big monitors, etc.

It connects to google voice recognition service which takes the speech and converts it to text and then sends it back to the client.

Note that the web service (see the demo page) accepts audio in the FLAC format.

This may help:

http://mikepultz.com/2013/07/google-speech-api-full-duplex-php-version/

http://updates.html5rocks.com/2013/01/Voice-Driven-Web-Apps-Introduction-to-the-Web-Speech-API

Web speech api is in very early stage. http://caniuse.com/#feat=speech-recognition

I think a solution can be frankensteined without a proper tinymce plugin for chrome at least.