I’m currently working on a Android+AppEngine project using voice as my main input method. On android your are able to use the speech package to translate the voice commands into plain text. The voice recognition isn’t done on the device itself, but send to a google server, which returns the text.
For your reference:
http://developer.android.com/resources/articles/speech-input.html
My goal is to do the voice recognition on the appengine side using the same google service. This means sending the voice recording to appengine storing it in blobstore and running the voice recognition service when nessecary. Is there a way to hook into the same service (preferably in java)?
I did some research and found some thirdparty solutions using i.e microsoft which might work. But before spending much time integrating such a solution i would like to find out if it is possible to stay in the google ecosystem.
AFAIK this API is not publicly available. This post on google groups discusses voice recognition: http://productforums.google.com/forum/#!topic/voice/YoUZStcv_EA
Here are some links where people have been successful in invoking Google speech API. Note that this is not an official API so there is no support and it might go away any time:
http://techcrunch.com/2009/12/14/the-unofficial-google-text-to-speech-api/
http://mikepultz.com/2011/03/accessing-google-speech-api-chrome-11/