Assignment Two

due Tuesday February 15th at 11:59pm

Assignment 2 is due February 15th (Tuesday) at 11:59pm.

Use the singing voice physical model and make "Sheila" sing a song. It does not have to be any real song at all, of course.

A few requirements, of course:

• duration: must last at least 20 seconds (and no more than 10 minutes :-)
• instruments: make use of singer.ins (see physical modeling).
• note events and/or phonemes sung: you should use celular automatas or fractals to control the phonemes being sung by "Sheila". The singer is controlled by a list of lists, where each sublist is a phoneme or transition between phonemes (see the web page for more details). You could automatically create a list of lists by using the "collect" feature of the "loop" macro. For example:

  (loop repeat 5 for i from 1 collect (list i))
  

  You have many parameters you can control, shape of the vocal tract, frequency, vibrato amount and so on and so forth.

  The "song" could be only one note, with all the vocalization controlled through the internal list of phonemes (no problems with Sheila ever having to breath :-), or it could be several notes, but in the later case I still want to see "internal control" of at least one note through a list of phonemes.

Your submission should include all the lisp code that generates the sound file and references to all instruments necessary to recreate the submission. Please test the code before including it in the submission.