This project attempts to demonstrate how easy it can be to do voice recognition and voice synthesis using any webkit enabled browser. At the moment though this does work on Firefox though it is based on webkit, but things will change soon.
Using a simple Angular.js module I demonstrate how simple it is to talk to an express server to return an answer for a given question and how the user can simply talk and listen.
There is also a server to simulate an AI replying automaton.
You could simply build a lookup table of strings and return something based on a pattern match. Possibilities are endless of course.
$ npm install
$ cd public
$ bower install
$ npm start
Go to http://localhost;2300
Girish Venkatachalam [email protected]