The Obstacles to Computer Lip Reading Technology

October 6, 2014

The article titled Lip Reading Is Still Too Hard for Computers on Engadget discusses the future of computer lip reading. Due to the complicated nature of lip reading, the technology is still far from a reality. A huge part of the problem is that so much of lip reading depends on conjecture. This is due to the fact that while humans make around “50 distinct sounds”, our mouths only form “between 10 and 14 distinguishable shapes.” Therefore, someone who is lip reading depends a great deal on cues beyond the shapes of sounds in the mouth. The article paraphrases the report of Ahmad Hassanat, who works at the Mu’Tah University in Jordan as a researcher,

“To suss out exactly what sounds a speaker is making, lip readers have to take in body language, facial expressions and the context of the conversation to help them decipher words. The researcher’s own experiments have produced an average success of 76-percent, but Hassanat says we still have a long way to go — in addition to missing out on contextual clues, he says, automated systems often fumble when reading the words of bearded men.”

Look out for more bearded men in the future; they may be trying to conceal their speech from eavesdropping computers. Whether you are excited or made nervous by this technology, it is clearly nowhere near complete.

Chelsea Kerwin, October 06, 2014

Sponsored by ArnoldIT.com, developer of Augmentext

Comments

Comments are closed.

  • Archives

  • Recent Posts

  • Meta