| Posted by: Zooped, March 5th, 2010 - No Comments » |
|
By Maggie Shiels

YouTube is making the tens of millions of videos it hosts more accessible to the deaf and hard of hearing by putting automatic captions on them.
The Google-owned company said this use of speech recognition technology is probably the biggest experiment of its kind online.
Previously captions were only on a small amount of content.
“A core part of YouTube’s DNA is access to content,” said the firm’s product manager Hunter Walk.
YouTube said by opening all this content to those who have not really been able to access it in the past should democratise information and “help foster greater collaboration and understanding”.
Initially the feature will apply to English language videos, with other languages being added in the coming months.
In November last year, YouTube rolled out automatic captions to a handful of partners including the University of California, Berkeley, Yale University and National Geographic.
‘Real solution’
The technology behind speech recognition has been around for about 50 years, said Google engineer Mike Cohen, and has finally become good enough to be used on a large scale.
“I have been working on speech technology for 25 years,” Mr Cohen told the BBC.
“There have been steady improvements and this is the culmination of lots of work over years and years. We have had to work on a wide variety of problems like accent variation, background noise, the variation in language, in pronunciation.”





