Lip Reading: A Modest Advance?
November 21, 2016
Watching surveillance videos without sound? Wish you could read lips? Don’t have time to learn how to read lips? Alphabet Google’s DeepMind has a solution for anyone in this predicament. “Google’s DeepMind AI Can Lip-Read TV Shows Better Than a Pro” reveals:
A project by Google’s DeepMind and the University of Oxford applied deep learning to a huge data set of BBC programs to create a lip-reading system that leaves professionals in the dust.
How accurate is the system? The write up states:
The professional annotated just 12.4 per cent of words without any error. But the AI annotated 46.8 per cent of all words in the March to September data set without any error. And many of its mistakes were small slips, like missing an ‘s’ at the end of a word. With these results, the system also outperforms all other automatic lip-reading systems.
I think this means that the Google method is almost four times more accurate. Software is faster and does not require health care, vacation days, and coddling.
The write up sidesteps law enforcement use of the system by emphasizing “improved hearing aids, silent dictation in public spaces, and speech recognition in noisy environments.”
There are other applications, however.
Stephen E Arnold, November 28, 2016
Comments
-
- Subscribe to Beyond Search
Feature archive
News archive
-
