In 1936 H.W. Dudley, a Bell Labs scientist, invented the first electronic speech synthesizer. Since that time AT&T Labs has been at the forefront in developing this technology. In 2001, AT&T unveiled the most advanced synthetic speech system to date, AT&T Natural Voices. At the heart of this technology is the AT&T Natural Voices Text-to-Speech (TTS) Engine, and this engine supports a library of multilingual male and female voice fonts in languages including U.S. English, Latin American Spanish, German, U.K. English, Parisian French and Canadian French (and this list will continues to grow). Also unique to AT&T Natural Voices are custom voice icons, exclusive voices that will provide a fast and effective way for businesses to extend corporate image, brand and personality.
AT&T Natural Voices' TTS technology is the key to giving voice-a pleasant, natural and crystal clear voice-to a new generation of AT&T managed business services. Integrated with other AT&T Labs speech technologies-including speech recognition, natural language understanding, and dialog management-Natural Voices is "Closest to the customer's ear," providing human-like speech output capabilities that will help accelerate the use of speech technologies in automated customer interaction systems.
Although not sold as a stand-alone product, AT&T Natural Voices has been licensed by companies and organizations seeking to make the Web accessible to visually and physically impaired users. Natural Voices is used by Freedom Box, a conversational Web interface that tells users the choices they have and describes exactly what is available on a Web page. Other text reading programs using Natural Voices include NextUp's TextAloud, and ReadPlease.