SignSynth, a Signed-Language Synthesis Application Using VRML and Perl

Document Sample
SignSynth, a Signed-Language Synthesis Application Using VRML and Perl Powered By Docstoc
					SignSynth, a Signed-Language Synthesis Application Using VRML and Perl
                                 Angus B. Grieve-Smith
                                 Linguistics Department
                              The University of New Mexico
                                     April 25, 2001

       Development of sign synthesis (also known as text-to-sign) can benefit from

studying the history of its older cousin, speech synthesis. Klatt (1987) outlines the basic

architecture of a speech synthesis application and makes the distinction between

concatenative speech synthesis, which rearranges prerecorded speech to make new

utterances, and articulatory synthesis, which builds an utterance from scratch.

       SignSynth is a CGI-based articulatory sign synthesis prototype under

development at the University of New Mexico. It takes as its input text a signed-

language text in ASCII-Stokoe notation (chosen as a simple starting point) and converts it

to an internal feature tree. This underlying linguistic representation is then converted into

a three-dimensional animation sequence in Virtual Reality Modeling Language (VRML

or Web3D), which is automatically rendered by a VRML browser.