Welcome to I.L.A. the Intelligent Learning Assistant!
is a voice activated personal assistant very similar to Apple's Siri, Microsoft's
Cortana or Google Now, but with the big difference that you
can teach it new commands by yourself and it is highly customizeable!
It is designed to run on your desktop or media center PC and integrates into your home enviroment with ease becoming very interesting for smarthome projects. ILA can search
the web for you, find locations, get directions, start timers and set reminders, read news headlines, open programs, execute system commands and much more ... just try it! It's especially fun when used with a bluetooth headset or microphone array to freely move around in your home while talking to ILA :-)
ILA is free and can be extended by everyone! It is programmed in Java and runs on Windows, Mac and Linux as long as you have at least Java RE 7 installed.
Right now ILA supports
the open source speech engines Sphinx-4 and Pocketsphinx, projects by
Carnegie Mellon University for offline speech recognition and a Google speech
API for cloud based online
speech recognition. ILA is designed modular to easily implement any new options/developements in the field of speech recognition.
Online (cloud based) recognition
is not required to run ILA but can greatly enhance the user experience because it is very robust, enables
support for many languages, needs basically zero resources (optimal for Raspberry Pi etc.) and performs fast, high accuracy, large vocabulary recognition where as the offline, open-source system works best with
small vocabulary or pre-defined sentences (auto-expanding while you teach ILA) and supports only a few languages so far. The drawbacks are that it is not customizable so it might not be super accurate when you ask about your contacts or individual information etc. and there is this uncertainty of what happens to your data in the cloud. Using the Google web speech API also requires to run Chrome or Chromium browser in the background.
Hope you enjoy ILA!- Florian