http://www.cslu.ogi.edu/toolkit
The freely available CSLU Toolkit includes a speech recognition engine and the
Festival text-to-speech synthesis system. Users can rapidly prototype spoken language
systems using a graphical authoring tool.
http://morph.ldc.upenn.edu/ltts/
The COCOSDA/LDC interactive TTS comparison site. Listen to the same text
synthesized by several commercial and academic synthesizers.
http://mambo.ucsc.edu/
UCSC Perceptual Science Laboratory with information on speech perception by ear and eye,
and facial information.
http://asel.udel.edu/speech/
Speech research at the University of Delaware. An online demo of ModelTalker is available
and capable of producing speech with a variety of emotions