Slashdot is carrying a post about Lip-Reading Surveillance Cameras
Now the British government is considering taking it literally by adding lip reading technology to some of the four million or so surveillance cameras in order identify terrorists and criminals by watching what everyone says.
Computer-based lip-reading technology would help video surveillance systems spot people planning a crime or terror attack by literally watching suspects’ lips for clues. Once it finds someone speaking certain key words or sentences, the system would automatically send an alert message to a central console, mobile phone, or other communications device. Police or security agents could then be dispatched to the scene to question the individual.
Lipreading is difficult enough for humans, thus how does the government expect a computer to do it? A case of artificial intelligence overtaking human intelligence? Has anyone bothered to figure how you lipread the difference between PAPER and BABY for example? Since criminal trials expect a standard beyond all reasonable doubt, how is this going to be admissible in court?
I've got a better idea. If this ever pulls off, how about trying this out on the next generation of speech to text captioning? Perhaps we could expand our access when it comes to accessing some of those spoken language vlogs we are currently unable to?

Comments (3)
FFS, that's a crap idea! Impossible. For example, someone seeming to say "Have you got the bomb and the matches?" may well be saying "Have you got the pop and the sandwiches?"
Ooh, room for crazy, crazy mistakes there.
Posted by j | May 1, 2007 8:57 PM
It still wouldn't help us here in the UK to follow ASL blogs.
Posted by MM | May 2, 2007 9:18 AM
It won't work. I have trouble lip reading never mind computers!!
Posted by Fintan | May 3, 2007 4:33 PM