voiceCmdr

voiceCmdr is a library for adding voice commands to your website. It is build on top of Webkit Speech API.

Browser support

• Google Chrome 31+

Getting started

Install with npm:

npm install voice-cmdr

Install with bower:

bower install voice-cmdr

Run http server (can be installed with npm install -g http-server):

http-server

Then open localhost:8080 in your browser and navigate to examples directory.

(*) Voice recognition works only through http protocol. That’s why you have to serve static files through the server instead of just opening them in the browser.

(**) Pages hosted on HTTPS do not need to ask repeatedly for permission, whereas HTTP hosted pages do.

API

• Add command:

    voiceCmdr.addCommand("command", callbackFunction);
  

  Example:

    voiceCmdr.addCommand("home", function () {
    	// navigate to home page
    });
    
    voiceCmdr.addCommand("search", function (param) {
    	console.log("Searching for:", param);
    });	
  

• Remove command:

    voiceCmdr.removeCommand("command");
  

• Start listening (continuously):

    voiceCmdr.start();
  

• Stop listening:

    voiceCmdr.stop();
  

• Get single command:

    voiceCmdr.getCommand();
  

• Turn debug mode on/off:

    voiceCmdr.debug(true|false);
  

Demo

www.bookslib.azurewebsites.net

Available commands:

books
favorites
top 10
search [string]

To Do

• lowercase commands
• Stop restarting after 5 tries (if user set deny the restarting will work infinitely)
• Add setting language
• Tests (???):
  • addCommand
    • Works
    • Can overwrite existing without error
  • removeCommand
    • Remove existing returns true
    • Remove non-existing returns false
  • in browser with karma (???)
  • Jasmine vs Mocha+Chai+Sinon with mocking window + webkitSpeechRecognition (???)

License

MIT