Automated Lip Reading

Frank Hubner has developed some rather clever tech which can produce a transcript of what someone is saying in a silent film. Image processing techniques are used on the video of their lips as they speak, this is then profiled against a database of recognised words to produce a transcript of what was said. This tech could be coupled with existing voice recognition techniques to give more accurate results.



In addition, this opens the way for automatic sub-title generation, and even automatically translated subtitles for all those Japanese films like 男たちの大和 I'd love to watch, but which presently don't have 英語字幕 (English subs).