Monday, 8 October 2012

Google Virtual Brain - A Self Learning System

Google is known for surprising the people and it has shown to continue it till today. Google’s breakthrough with its latest virtual brain technology in the field of AI intends to identify animals, people and other objects by simple watching videos. This self-learning system is surely one of its kinds if implemented and can benefit a lot with its speech recognition facility.

The new virtual brain technology basically works on prompting the brain cells that communicate and react to each other responses. This neural network between the cells lets the brain to response at an event occurred and hence the brain is said to have learnt something.

Not only speech recognition is incorporated, but the amazing virtual brain also works on pattern or context matching. This clearly means that if someone would say “I am having a banana!” and somehow the sentence gets a bit stifled then the system would work on the basis of guess or context matching by suggesting the other things in the same domain for instance, “Mango”, “orange”, “apple” etc. (as matching the domain of the example given above).

This innovative technology can be of great significance to Google’s smartphone OS Android because of its excellent speech recognition ability. The virtual brain is claimed to work on US English perfectly by Google’s speech recognition experts and is currently under process to identify other dialects and languages as well.

Undoubtedly, Google has succeeded in revolutionizing the world with its novel ideas like virtual brain smart learning system. According to Yoshua Bengio, an AI expert at University of Montreal, “This is simply a path towards making more general artificial intelligence systems”.