DIASS - Demonstration Sounds

Here are some demonstration sounds that were produced with DIASS on Argonne's IBM SP. The visual image gives an idea of the complexity of the sounds. You may download the sound files (.snd or .aiff format) and play them at a sampling rate of 22,050Hz.

  1. The “ARGONNE Chime”
  2. Loudness of Sounds
  3. Loudness of Sound Clusters
  4. Detuning
  5. Detuning-Tuning
  6. Dynamic Control
  7. Reverberation
  8. Sound Movement

  1. The “ARGONNE Chime”

    The "Argonne chime" consists of seven sounds,

    spelling A R G O N N E
    in musical sounds a re (= d) g sol (= g) (percussive) e (low gong)

    Six sounds have six partials (sine waves), one sound has sixteen partials. The sounds have different degrees of vibrato and tremolo and distinct frequency and amplitude transients. Individual envelopes vary from percussive (sounds 5 and 6) to sustained (sound 7). Various degrees of reverberation and channel placement (left, right, center) create a realistic environment.

    chime.snd (1.1 MB) or chime.aiff (0.5 MB)


  2. Loudness of Sounds

    The perception of loudness depends on the frequency and the timbre (number of partials) of the sound. This demo serves to illustrate the concept of perceived equal loudness. It has sounds with 4, 16, and 64 partials, which are repeated at frequencies of 120, 480, and 2100 Hz. Each sound is played at successively increasing loudness levels of 40, 60, 70, 80, 85, and 90 phons. (A phon is a measure of loudness.) If your speakers respond evenly across the frequency spectrum, sounds of equal loudness will be perceived as being equally loud.

    The loudness routines in DIASS incorporate various results of psychoacoustic research. The software uses the Fletcher-Munson curves of equal loudness and the concept of critical bands as formalized by Stevens and Zwicker, Flottorp, and Stevens.

    loudness.snd (6.4 MB) or loudness.aiff (3.2 MB)


  3. Loudness of Sound Clusters

    This demo again serves to illustrate the concept of equal loudness across the frequency spectrum and for different timbres. It has five sound clusters, each lasting approximately 5.5 seconds. The sound clusters have 24, 5, 15, 1, and 10 sounds, respectively, with frequencies ranging from 40 to 5,000 Hz. The number of partials in these clusters is 754, 113, 453, 60, and 250, respectively. (The maximum number of partials per sound is 60, the cutoff frequency for the partials is one-half the sampling rate of 22,050 Hz.) All partials are assigned the same amplitude, which presents the worst case scenario when one tries to obtain the same perceived loudness for all clusters. Each cluster has a loudness of 90 phons. (As in the previous demo, the 40 Hz sound, which is the only sound in the fourth cluster, may seem softer because of a particular speaker's response at that frequency.)

    cluster.snd (2.5 MB) or cluster.aiff (1.2 MB)


  4. Detuning

    Sounds with 180 Hz fundamental and 4, 16, and 64 partials start out by being in tune, then get more and more out of tune. DIASS uses a stochastic algorithm in the detuning process. The average rate at which partials are detuned is specified by the user.

    tuning.snd (1.8 MB) or tuning.aiff (0.9 MB)

  5. Detuning-Tuning

    The demo starts with one sound (180 Hz) corrrectly tuned. The sound is progressively detuned. A second sound (also at 180 Hz) with inharmonic partials overlaps with the ending of the first one and is progressively tuned.

    de_tune.snd (1.8 MB) or de_tune.aiff (0.9 MB)

  6. Dynamic Control

    This demo contains an old melody, which is made out of eight groups of five sounds. The sounds demonstrate how the magnitude (sounds 1-5, 11-15, 21-25, and 31-35) and the rate (sounds 6-10, 16-20, 26-30, and 36-40) of tremolo (sounds 1-10), vibrato (11-20), amplitude transients (21-30), and frequency transients (31-40) can be controlled.

    old_tune.snd (6.0 MB) or old_tune.aiff (3.0 MB)

  7. Reverberation

    This demo illustrates the reverberation of sounds in rooms of various sizes. The first sound has no reverberation at all. Sounds 2, 3, and 4 give the effect of a small room, sound 5 of a medium-size room, sound 6 of a large room, and sound 7 of a very large room. The length of the sounds increases from 2 seconds (sounds 2, 3, and 4) to 5 seconds (sound 7).

    reverb.snd (3.0 MB) or reverb.aiff (1.5 MB)

  8. Sound Movement

    By manipulating the reverberation, panning, and perceived loudness one can create the illusion of movement in space. This demo illustrates the effect on 34 low percussive sounds. The source moves from front-center to left-middle ground to center-back to right-middle ground and then back to front-center.

    walking.snd (6.2 MB) or walking.aiff (3.1 MB)

    Back to DIASS

    Last update: October 21, 1998 (HGK)