Advances in medical speech recognition software has made the technology more
accurate than ever, but certain weaknesses remain, radiologist David Weiss said
at the recent Radiological Society of North America 2010 conference in
Chicago.

Weiss said the sound of a closing door can be a “disaster” for speech recognition, but he’s found a quick-fix remedy that’s as easy as flicking a
switch. While white noise generators can effectively drown out garbled
background noise, Weiss said he uses a fan.

Weiss said the fan creates enough white noise to work as effectively as a
generator in making it easier for the software to recognize and translate
speech, but for much less.

The doctor also identified several other tricks he employs to boost the
benefits of his speech recognition platform. For example, Weiss uses a macro
tool for commonly dictated phrases that recognizes single words to enter entire
clauses.

He also praised the technology for its high accuracy, noting how the software
may actually “learn” the voice patterns of the dictating physician.

Healthcare organizations worldwide are utilizing speech recognition software
and transcription services to increase their efficiency and productivity.
Dictation products for companies in all industries can be found at the comprehensive American Dictation Site.