The E-Speech text-to-speech synthesizer serves
as the voice of the popular reverse telephone directory service now
available in Illinois, New Jersey, Maryland, and coming soon to other states. In an
independent telephone-based experiment, names and addresses were understood more often
when pronounced by the E-Speech TTS system than by other commercial synthesizers, and they were transcribed
more accurately, in less time, and with less effort, as shown in these figures.
Source: Silverman, K. Kalyanswamy, A., Silverman, J.,
Basson, S., and Yaschin, D.
"Synthesizer intelligibility in the context of a name-and-address information
service," Eurospeech, 2169-2172, 1993.
|
