Music 448

Sound Maker

Files need to run:

To use Sound Maker you need to have the executables, the library, and the script model in your directory:
cp /home/sever/Classes/M448/Smaker/sm .
cp /home/sever/Classes/M448/Smaker/sm.libr .
cp /home/sever/Classes/M448/Smaker/smR .
cp /home/sever/Classes/M448/Smaker/smS .
cp /home/sever/Classes/M448/Smaker/scri15 .
Be careful and make sure that some words are capitalized.
The "." at the end means you want to copy those files in the directory you are in at the moment.

Terms used and comments

Start time: attack time, recommended that at least one sound has a stime of 0. To create chords, give the same stime to all elements.
Partials: same as overtones; SoundMaker uses sine waves, the simplest type of wave. More partials will help produce a more complex, possibly more appealing sound.
Frequency: given in Hertz or cycles/second, tuning A₄ = 440 Hz. Possible range is 20-12,000 Hz. but most speakers will not render well sounds bellow 35 Hz. or so. Also, remember that our percption of pitch (related to frequency) uses a logarithmic scale: A₃ = 220, A₄ = 440, A₅ = 880, A₆ = 1760, A₇(the highest on the piano keyboard) = 3520.
Loudness: is given in sones also a scale using powers of 2. It roughly corresponds to 256 -> ff, 128 -> f, 64 = mf, etc.
Deviation: a number between 0 - 1, a probability. Partials are integer multiples of the fundamental:
```
		partial			Hz.
		  0 (fundamental)	220	1 * 220
		  1			440	2 * 220
		  2			660	3 * 220
		  3			880	4 * 220
		  4		      1,100	5 * 220
	etc.
```
When deviation = 0, partials will be in the above relationship; when deviation = 1, partial 1 and higher (not the fundamental) can have a value between half the distance to the lower neighbor and half the distance to the upper neighbor actual value being chosen randomly.
Envelopes and scaling factors for partials: define the evolution of amplitude (corresponding to perceived loudness) during the duration of the sound. To see these envelopes, use gedit or your favorite text editor to open the file sm.libr. Each envelope has an identifier number (eg. Envelope 9); second line shows how many segments are in this envelpe (eg. 5); the first column contains the x values and the second column the y values:
```
          x      y
	0.00    0.00     LINEAR         FLEXIBLE        0.486
	0.43    1.00     LINEAR         FLEXIBLE        1.14
	1.57    1.00     EXPONENTIAL    FLEXIBLE        0.486
	2.00    0.00
```
the third column shows the path: a straight line or an exponential curve; the fourth columns specifies if the segment has a fixed value or if it can be stretched or compressed in time; last column shows the length/duration of the segment. The scaling factor, multiplies every point on the envelope: (eg. 0.5 will reduce the y by half).
ATTENTION: envelopes used to control partials amplitude HAVE TO stat and end with y = 0 otherwise there will be a "pop" or distortion at the beginning or at the end of the sound.
If you add new envelopes remember to change the first number (on the first line) otherwise the program willnot read your new entries.
Modifiers: modify the frequency and/or the amplitude (FM or vibrato, AM or tremolo, Glissando also a change in frequency).
- Both vibrato and tremolo require two sets of envelopes with their scaling factors, one for the size/magnitude or amplitude of the effect, the other for the rate of change.
  Envelope numbers (called from the library sm.libr are integer numbers while scaling factors are numbers between 0-1 for the size/magnitude/amplitude of the effect - it is recommended that a small value is used (eg. 0.1).
  Scaling factors for the rate are larger numbers (eg. a "normal" vibrato has a rate of change of 6-8 times/sec.). Smaller numbers (eg. 1 or 0.5) will produce a "siren" type of effect. Much larger numbers will create side bands and an effect similar to the FM synthesis technique.
- Glissando requires only one envelope and scaling factor. This feature also modifies the original frequency by multiplying the scaling factor with each point of the chosen envelope. When y = 1 the fundamental frequency remains constant, when y = 0.5 the frequency drops one octave and when y = 2 the resulting freauency is one octave higher.
Panning determines the location of the sound in the stereo field and requires an envelope and a scaling factor. When y = 0 the sound will appear to be produced in the left speaker; when y = 1, in the right speaker; a value of 0.5 will crate the illusion that the sound comes from half way between the two speakers.
Similar to the other modifiers, using Envelope 1 will produce a constant effect while Envellpe 2 and Envelope 3 will move the sound from one speaker to the other over the duration of the sound. Use the last two envelopes on sonds that are 1 sec. or longer for an obvious result.
Reverberation: room size = 0 will produce a "dry" sound (no reverberation) while room size = 1 will create the illusion of a very large hall (max. reverb).
Sound file: name your soud file adding .aiff (eg. mySound.aiff) otherwise Audacity or other applications will not recognize your audio file as such.

Music 448

Sound Maker

Enjoy !