|Weather Radio Gets a New Voice|
After reviewing over 19,000 Internet comments, NOAA has selected new "voices" for its Weather Radio automated radio weather warning system. Say goodbye to Paul, weather buffs. Stand by for the Donna and Craig Show.
Since he first went on the air in 1997, millions of listeners to the National Weather Service's (NWS) weather warning system have become familiar with "Paul," the original NWS computer-synthesized voice. NWS upgraded the warning radio system in 1997 to utilize a computer-generated voice in order to enabled the weather service to send out multiple independent warnings over multiple transmitters simultaneously, allowing speedier delivery of severe weather warnings and more lead-time for the public.
While he had become one of the most-heard voices in the history of radio, Paul really did sound like a computer and could be a little hard to understand with with a gusty thunderstorm, tornado or hurricane swirling around you.
"The old voice was state-of-the-art when first placed in service in 1997, but advances in artificial speech technology now make it possible for us to provide a service that is more understandable to the public," stated retired General Jack Kelly, director of the National Weather Service in a recent press release. "This reinforces our commitment to continuously use new technology to improve the timeliness and effectiveness of public warnings."
After deciding to retire Paul, NOAA awarded Siemens Information and Communication Network of Boca Raton, Fla., a $633,615 contract to develop improved computer voices. As part of the contract, Siemens teamed with SpeechWorks International of Boston, Mass., to provide software that combines phonetic sounds with natural language modeling.
"SpeechWorks is working with us to make NOAA Weather Radio even better, said Greg Mandt, director of the weather service's Office of Services. Mandt said a series of focus groups conducted around the country are indicating wide support for the new voice. "We'll use this feedback to help us further improve the voice and find out what changes our listeners might want in the NOAA Weather Radio service. We're confident these voice improvements will help to increase NOAA Weather Radio's listening audience, which translates to protecting more lives and property."
Based on input from listeners, NOAA selected new male and female voices -- Craig and Donna -- to replace Paul beginning early in 2002.
"These voices are more understandable and human-sounding than the current voice, and will help NWS to deliver warnings, watches, forecasts, and other hazard information more quickly and accurately," states the NWS.
NWS encourages visitors to use Windows Media Player® to listen to the voices, if at all possible, for consistency. If you do not have that player on your PC, you may want to download RealPlayer® as an alternate. Click here for a free copy of RealPlayer®.
NOAA Weather Radio, sometimes referred to as the voice of the National Weather Service, is a portable device that enables the public to receive continuous weather broadcasts and hazard alerts directly from local weather forecast offices. Transmitting from a network of 583 stations nationwide, the NOAA Weather Radio can be heard by more than 85 percent of the U.S. population.