My programming skills are several decades stale. I need help developing a demo and research tool. It involves modifying the downloadable code for an existing tool for speech synthesis so as to admit disturbances to acoustic parameters of the synthesized speech signal.
The heavy lifting was done in the late 1970s by Dennis Klatt. His program for speech synthesis is still very much in use. Here is the original (Klatt 1980) paper:
http://www.fon.hum.uva.nl/david/ma_ssp/doc/Klatt-1980-JAS000971.pdf
And a brief wiki page:
http://linguistics.berkeley.edu/plab/guestwiki/index.php?title=Klatt_Synthesizer
Several variants and descendants of the Klatt program are available for free download. This refinement by one Bob McMurray at U. Iowa seems to me the best for PCT research. It provides a lot of very useful flexibility.
http://www2.psychology.uiowa.edu/faculty/mcmurray/klattworks/default.html
___________________ <
/div>
For background and some hands-on play opportunities, here’s a collection of web interfaces put up by another person using the original Klatt program (with a slight modification that is inconsequential for what we’re looking at).
Demo in which static values may be set for F0, F1, F2, and F3:
http://www.asel.udel.edu/speech/tutorials/synthesis/vowels.html
More elaborate
web demo that exposes more variables:
http://www.asel.udel.edu/speech/tutorials/synthesis/Klatt.html
The variables are documented at the end of the ‘man page’ for the underlying gensyn program:
http://www.asel.udel.edu/speech/tutorials/synthesis/gensyn.htm
These web interfaces are a useful introduction to the acoustic variables involved in speech recognition. They could be used to demonstrate that distinct combinations of formant values are r
ecognized as different vowels. The program and code are not offered for download.
http://www.asel.udel.edu/speech/tutorials/synthesis/Klatt.html
In any case, a dynamic interface is needed to demonstrate the effects of varying the formant values. I believe McMurray’s program above offers this.
···
/Bruce