K404/K504 Assignment 4: More Sound File Processing
In this assignment, we'll look at four interesting Plug-Ins:
- RiverRun, a Digital Performer plug-in that performs real-time
granulation of sampled sound
- Spektral Delay, a Digital Performer plug-in that creates a range
of fascinating delay effects
- Periscope, a Digital Performer plug-in that lets you create
complex equalization curves (review from last semester)
- MasterWorks Compressor, a Digital Performer plug-in for sound
compression
The goal of the assignment is just to get a feel for each of these tools, and
to make some sound that interests you. As always, try to imagine how you might
use these sounds in your final project.
Granulation with RiverRun
First, we'll explore a Digital Performer plug-in, called RiverRun, that
performs real-time granulation of sampled sound. To granulate a
sound is to break it up into many tiny grains — each anywhere from
a few milliseconds to about a half second in duration. Each grain is encased
in an envelope and can be independently transposed. The grains flow out of the
granulator in one or more streams, one grain after another. The time between
successive grains determines the grain rate, or grain speed,
which has a profound impact on the sound. If the grain rate is regular, you
may hear a pitch with a frequency corresponding to the time between successive
grains. If the grain rate is randomly varied, this pitch disappears.
Why would anyone want to granulate sound? The applications range from
time-scaling audio to making a sustained sound out of a short snippet (e.g., a
choral sustain from a guitar pluck) to creating interesting noisy or pulsed
repetitive textures. Play with it and see.
Granular synthesis does the same thing, but instead of taking grains
from a sound file, it synthesizes grains, using simple wavetable or FM
synthesis. (We don't have a plug-in that does this.)
One of the first composers to explore granular synthesis, Barry Truax (from
whose composition, Riverrun, the plug-in name, comes), likes to use
granulation as an audio microscope, zooming in on sound and exploring its
details in slow motion.
After you've played with RiverRun a bit, it'd be a good idea to read a more
complete description of granular synthesis, such as the one provided in
Computer Music Tutorial, by Curtis Roads (on permanent reserve for
Hass).
Running RiverRun
I'll provide a few brief instructions to help you get started, but you
have to experiment to really learn this plug-in. In the Documentation
folder, there is a manual for the "Nautilus Bundle," which includes two
other plug-ins besides RiverRun. Read the RiverRun section.
When you assign RiverRun as an insert for an audio track, you see the following
window.
RiverRun works differently from other plug-ins, because it requires that you
capture some sound into its waveform display. Then the sound is frozen, and
you use the controls to traverse and granulate the sound. At that point, you
don't even need to play the sequence — RiverRun always puts out sound,
unless you bypass it.
Here's how to get started.
- In Digital Performer, create a mono audio track. Make sure that
the track's output is stereo — it should read Analog 1-2.
- In the Mixing Board window, choose AudioEase > RiverRun
as an insert effect for your track.
- There are two ways to capture sound into RiverRun. RiverRun can
hold only about 3 seconds of audio.
- Press the load button in RiverRun, and select a mono
sound file in Sound Designer II format. (That's the
only format RiverRun can open.)
- Put a mono soundbite into the track that uses RiverRun.
Press the RiverRun record button (the big red
bull's eye), and then play the sequence. As the soundbite
plays, its waveform enters the RiverRun display from the
right. Press the RiverRun record button again to freeze
the segment you're interested in. Stop the sequence.
- Now you can play with the sound. To tell RiverRun: "Please be
quiet for a minute," use the Bypass button.
- Here are some problems to solve. When in doubt, read the manual
(in the "Manuals and Demos" folder on your Desktop).
- How do you stretch a sound so that it lasts four times as
long? What settings give the most natural-sounding result?
Try different grain envelopes, using the shape
pop-up menu.
- How do you play a sound backwards?
- How do you creep through a sound file so that you can hear
its timbral evolution in extreme slow motion?
- How do you mangle the input sound into radio static?
(Hint: Grains must be short, fairly sparse and played with
irregular timing.)
- How do you make a long sustained sound out of a short sound?
- How do you make it sound like a swarm of angry bees? (Hint:
Try random pitch variation.)
- How do you make the bees sing a major seventh chord. (Hint:
Pitch grid.)
- How do you make a pulsing repetitive groove?
Saving RiverRun output
You can use RiverRun "live" in your sequence, but at some point you'll need to
capture the sound it makes into a sound file. There are two ways to do this.
- Select a portion of the track that uses RiverRun, and choose
Bounce to Disk from the Audio menu. Unfortunately,
this doesn't seem to work as reliably as the next, more involved,
method.
- Record the output of RiverRun into another track.
- Create a new stereo track, and record-enable it.
- Set its input to New Stereo Bundle > bus 1-2.
- Set the output of Riverrun's track to bus 1-2.
- If RiverRun's track has a soundbite in it, delete the soundbite.
- Rewind and record. (Press the Digital Performer record button,
not RiverRun's record button.)
While recording, you won't hear anything, but you'll see the
waveform being written to your stereo track.
- Press stop, and do whatever you want with the soundbite that
appears in the record-enabled track.
Spectral Processing: Spektral Delay
Spektral Delay is a plug-in for Digital Perfomer that lets you create
all kinds of recirculating delay effects. Spektral Delay performs real-time FFT
analysis and resynthesis of the audio in a track. It lets you manipulate the
analysis data, prior to resynthesis, using the familiar model of a
recirculating delay line (i.e., a delay line with feedback). The difference is
that there is a separate delay line for each of the FFT frequency bands.
You control the amount of delay and feedback for each band individually. It
even has LFOs that modulate these values. You can also attenuate the amplitude
of the signal for each band before it enters its delay line (performing an
FFT-based EQ).
This is a big, complicated plug-in. Fortunately, the layout of the plug-in
window helps you to visualize the signal flow, from left to right.
- The audio enters at the left, split into left and right channels. The
left-most box displays a sonogram of the input, right after the
FFT analysis is performed. Frequencies are graphed on the vertical
axis; stronger frequency bands have brighter color.
- The box to the right lets you perform various canned spectral
modifications of the FFT analysis data.
- The next box, labeled FILTER, lets you draw an attenuation curve across
the frequency bands.
- The next box lets you draw the delay time for each frequency band.
Click the mouse in the graph, and as you drag, you'll see the FREQUENCY
and DELAYTIME readouts change to reflect your cursor position.
- The next box lets you set the amount of feedback for each delay line.
- The last box shows you a sonogram of the output, before resynthesis of
the modified FFT data.
Some tips:
- Click the FILE button (at the top) to see a bank of presets
(actually, 4 banks of 16 presets each, accessible via the A, B, C and D
buttons). Try out some of these presets. The pop-up menu above the last
column of presets lets you choose other preset files. Try these.
- If Spektral Delay still makes sound after you stop the sequencer
transport, you can silence it by clicking the trumpet-in-a-red-circle
icon one or more times.
- Hold down the option key while dragging in a graph to set all the
frequency bands to the same value. Hold down the apple key to shift all
the bands by the same amount.
- If you want to capture the output of Spektral Delay into a sound file,
solo just your Spektral Delay track, select a time region, and choose
Bounce to Disk from the Audio menu. Be sure to select
enough "empty" time (following the end of the sound clip) to allow the
Spektral Delay output to decay. If you turn feedback all the way up for
many frequency bands, it will never decay.
The Spektral Delay manual is in the "Manuals and Demos" folder.
Graphic EQ: Periscope
Periscope is a Digital Performer plug-in that performs a real-time
FFT analysis on the audio in a track and lets you change the amplitudes of
the FFT frequency bands before resynthesis. This amounts to a very
flexible and precise graphic EQ.
In the Mixing Board window, choose AudioEase > Periscope as an
insert effect for a track.
The top portion of the Periscope window gives you 32 frequency bands of
EQ, with sliders to control the amount of boost or cut for each band. The
bottom portion of the Periscope window lets you adjust the frequency range
within which the 32 sliders operate. Drag the red and yellow needles
around to set a range. Notice that the calibration of the sliders changes
to show just frequencies within that range. Sliders on the left and right
ends are low and high shelf EQs, respectively. Sliders in between are
peak/notch EQs.
The fader edit mode buttons at the bottom right of the plug-in
window control how your mouse movements affect the positions of the
sliders. Free draw lets you, uh, draw a curve freely.
Magnetic lets you drag one slider while its neighbors move to form
a smooth curve. The magnetism knob controls the smoothness of the
curve.
You also see, underneath the sliders, a real-time graph of the sound,
post-EQ. (Actually, there are two graphs: the green graph shows peak
amplitude, while the blue graph shows average amplitude.) This helps you
evaluate the strength of frequency bands.

Play around to see how dramatically you can change the sound. Unlike the
Perigraphic EQ plug-ins, Periscope lets you make very steep, "brick wall,"
EQ shapes.
The manual is in the "Manuals and Demos" folder.
MasterWorks Compressor
MW Compressor is a Digital Performer plug-in that offers three separate
compressors with crossover controls, allowing you to set the frequency range
and compression characteristics of each band. Each frequency band can be
soloed or muted, allowing you to audition them separately.
Here's what some of the controls do.
- Threshold: Determines the level above which the compressor will
have an effect.
- Ratio: Determines the amount of compression on the signal over
the threshold.
- Makeup Gain: Adds Gain to the signal after compression.
- Attack: Determines the time it takes for the compressor to react.
- Release: Determines the time it takes for the compressor
attenuation to return to zero.
©2008, John Gibson, Christopher Cook