JCHOR -- granulated chorus instrument
in RTcmix/insts/jg


quick syntax:
JCHOR(outsk, insk, dur, indur, inmaintain, pitch, nvoices, MINAMP, MAXAMP, MINWAIT, MAXWAIT, seed, inputchan, AMPENV, GRAINENV)


   p0  = output start time (seconds)
   p1  = input start time (seconds)
   p2  = output duration (seconds)
   p3  = input duration (seconds, not input end time)
   p4  = maintain input duration, regardless of transposition (1: yes, 0: no)
   p5  = transposition (oct.pc)
   p6  = number of voices (minimum of 1)
   p7  = minimum grain amplitude (relative multiplier of p13)
   p8  = maximum grain amplitude (relative multiplier of p13)
   p9  = minimum grain wait (seconds)
   p10 = maximum grain wait (seconds)
   p11 = seed (0-1)
   p12 = input channel
   p13 = overall amplitude (relative multiplier of input sound)
   p14 = grain envelope table (pfield-handle)

   p7 (min amp), p8 (max amp), p9 (min wait), p10 (max wait) and p13 (amp mult)
   can receive dynamic updates from a table or real-time control source.

   p14 (grain envelope) should be a reference to a pfield table-handle.

   Author:  John Gibson, 9/20/98, RT'd 6/24/99; rev for v4, 7/24/04
       based on Paul Lansky's CMIX chor instrument  and Doug Scott's TRANS code



JCHOR is a random-wait chorus instrument based on Paul Lansky's old CMIX chor instrument and Doug Scott's TRANS. Essentially this is a granular synthesis instrument optimized to produce 'choir-like' sounds. Paul Lansky used this approach in man of his Idle Chatter pieces (the chords in the background).



Usage Notes

Because the transposition method doesn't try to maintain duration -- it works like the speed control on a tape deck -- you have an option about the way to handle the duration of the input read: JCHOR places no limit on input duration or number of voices. It does transpose the input signal according to the oct.pc value given in p5 ("pitch").

p14 ("GRAINENV") is a pfield-handle reference to a table with the envelope to be applied for each grain. NOTE: You will probably need to set the reset control-rate value (see the reset scorefile command) fairly high for small grains.

Output can be either mono or stereo. If it's stereo, the program randomly distributes the voices across the stereo field.



Sample Scores

very basic:

   rtsetparams(44100, 2)
   load("JCHOR")
   
   rtinput("mysound.aif")
   inchan = 0
   inskip = 0.20
   
   outdur = 10
   
   indur = 0.60
   maintain_dur = 1
   transposition = 0.07
   nvoices = 50
   minamp = 0.01
   maxamp = 1.0
   minwait = 0.00
   maxwait = 0.30
   seed = 0.9371
   
   amp = 0.5
   env = maketable("line", 1000, 0,0, 1,1, outdur-1,1, outdur,0)
   
   grainenv = maketable("window", 1000, "hanning")
   
   JCHOR(0, inskip, outdur, indur, maintain_dur, transposition, nvoices,
      minamp, maxamp, minwait, maxwait, seed, inchan, amp * env, grainenv)


slightly more advanced:
   rtsetparams(44100, 2)
   load("JCHOR")
   
   rtinput("mysound.wav")
   inchan = 0
   
   outdur = 16
   inskip = 0.01
   indur = 0.20
   maintain_dur = 1
   transposition = 0.02
   nvoices = 60
   minamp = 0.1
   maxamp = 1.0
   minwait = 0.00
   maxwait = 0.60
   seed = 0.9371
   
   amp = 6.0
   env = maketable("line", 1000, 0,0, 1,1, outdur-3,1, outdur,0)
   
   grainenv = maketable("window", 1000, "hanning")
   
   reset(2000)
   
   outskip = 0
   JCHOR(outskip, inskip, outdur, indur, maintain_dur, transposition, nvoices,
      minamp, maxamp, minwait, maxwait, seed, inchan, amp * env, grainenv)
   
   outskip = 2
   outdur -= outskip
   amp = 0.6
   indur = 0.90
   transposition = -0.10
   maxamp = 0.5
   maxwait = 1.0
   seed = 0.2353
   JCHOR(outskip, inskip, outdur, indur, maintain_dur, transposition, nvoices,
      minamp, maxamp, minwait, maxwait, seed, inchan, amp * env, grainenv)



See Also

GRANSYNTH, GRANULATE, JGRAN, SGRANR, STGRANR