AT&T Spearheading Speech Recognition Technology

Researchers at AT&T Labs are working on the latest technology when it comes to speech recognition, according to a recent CNET report.

The latest AT&T Labs innovation is the Watson engine, a cloud-based speech recognition platform. With Watson, researchers are hoping to advance beyond a system that recognizes mere words, but also interprets their meaning and emotions attached. According to CNET, with more data, Watson can evolve to improve accuracy.

"We are really on the cusp of a technology revolution in speech and language technology," said Mazin Gilbert, executive director of speech and language technology at AT&T Labs. "It's no longer about simply trying to get the words right. It's about adding intelligence to interpret what is being said and then using that to apply to other modes of communication, such as text or video."
 
Improved accuracy is a main goal for speech recognition software vendors and the enterprise users that take advantage of the technology's benefits. Innovation that leads to more accuracy could make the technology more useful for medical and legal professionals looking to improve their productivity and efficiency.

While improved speech recognition technology is desirable for some interested in improving efficiencies and reducing costs, experience has taught us that, for others, the personal interaction and service of a professional transcriptionist is still irreplaceable.


Where speech recognition can fall short for some is in the formatting and interpretation of dictations and legal documents.  An experienced transcriptionist has the advantage of knowing exactly what the provider or attorney is looking for as far as terminology and format, and knows when to eliminate extraneous words and utterances, something speech recognition has yet to perfect.