MIDI voice systhesis

http://cslu.cse.ogi.edu/tts/flinger/

Well, this is pretty damned cool. Have you guys seen this?

"Overview
Flinger is a program for synthesizing singing voice from a MIDI file
input. It based on

* the Festival Speech Synthesis System, developed at the
University of Edinburgh
* the C++ library contained in earlier versions of the TclMIDI
package, developed by Mike Durian (another TclMIDI page is here) "This
product includes software developed by Michael B. Durian."
* various modules developed here at the Center for Spoken Language
Understanding at OGI, including the OGIresLPC synthesizer and voices, and
* a singing-voice synthesizer written by Mike Macon with support
from Fluent Speech Technologies.

Flinger is similar in functionality to an earlier system called
Lyricos, but in no way uses any source code or data contained in the
Lyricos system.

Currently, Flinger is provided in pre-compiled form only, for Linux
and Windows platforms only. We plan to offer a source code
distribution in the future, similar to the way the other OGI TTS group
software is provided.
Flinger Jukebox
Visit the Jukebox page to hear some samples of a growing library of
music contributed by Flinger users around the world.
System Documentation
The Documentation page provides the most basic information needed to
use the system. Note that this is not "pointy-clicky" software. In
order to use it, you must be able to edit a text file and run a
command-line program from the MS-DOS (or Linux) shell.
References
Some publications describing the concatenative approach to singing
voice synthesis are as follows:

* M. W. Macon, L. Jensen-Link, J. Oliverio, M. Clements, and E. B.
George, "Concatenation-based MIDI-to-singing voice synthesis,'' 103rd
Meeting of the Audio Engineering Society, New York, 1997.
* M. W. Macon, L. Jensen-Link, J. Oliverio, M. Clements, and E. B.
George, "A system for singing voice synthesis based on sinusoidal
modeling,'' Proc. of International Conference on Acoustics, Speech,
and Signal Processing, Vol. 1, pp. 435-438, 1997.
* M. W. Macon, Speech Synthesis Based on Sinusoidal Modeling, PhD
thesis, Georgia Institute of Technology, October 1996.

A large number of references to other papers on singing voice
synthesis can be found in the last reference above.
Download
We are currently allowing a small number of qualified users to
beta-test Flinger for non-commercial use. In order to apply to be a
beta-tester, you must register. Your application will be considered
and if access is granted, a confirmation will be emailed to you. Your
name, address, email, etc., will not be sold or used for any other
purpose than this. To obtain the beta version of Flinger, you must

* Read the license agreement and fill in information on the
Registration Page.
* Wait for a confirmation email (usually within a week)
* Access the password-controled (use username and password you
have entred in registration form)Downloads Page

Support
We are looking for funding to continue Dr. Macon's work on Flinger.
However, in the meantime, we have no resources to fix problems or add
functionality. We have set up a forum at
http://speech.bme.ogi.edu/forum to provide Flinger users with a forum
to share experiences and trouble solving information. Please check
back here regularly after you've become an accomplished Flinger user
to pass on your knowledge."