reCAPTCHA uses the human intelligence to help the world of OCR. People solve not one, but two captchas, one is the one computer knows the answer to, the other one computer does not know. So answer only one is good enough if you know which one to answer. The other one will assume to be true if one of the captcha's result is correct.

It provide both visual and voice captcha. The voice captcha should only show the user the captcha the computer already knew, because it is not possible for computer to create voice out of a word it don't understand.

There is always a chance that a person got one of the word right the other one wrong, reCAPTCHA have a build in system just to prevent this. After a word is "known", the system will then show the word to other people to confirm if it's right or not.
I'm definitely using it soon :)