Is it possible to use the synthesised speech from Web Speech API as a SourceNode inside Web Audio API\'s audio context?
You can use Google's Web Speech API, you record the sound on your local machine and it is send to an external server, you can control some variants like when to stop or start the recognition and some other things. For more information here's a link:
http://updates.html5rocks.com/2013/01/Voice-Driven-Web-Apps-Introduction-to-the-Web-Speech-API