[music-dsp] [OT] FW: Alien Project track: My alien is GAY!

Sarah Thompson sarah at telergy.com
Sun Feb 27 16:14:15 EST 2000


Hi all in dsp land...

I've recently finished off a for-fun sound design challenge that started on
the music-bar at teklab.com list recently. The basic idea was that a sample of
someone saying 'Feed the Aliens' was put up on the web, with list members
being challenged to turn it into a track. Not being one to pass up a
challenge, I hit the problem with a large, Kyma-shaped hammer. The result is
pretty silly, but in case anyone out there hasn't heard Kyma used in anger,
I thought it might be of interest.

Sarah

-----Original Message-----
From: Sarah Thompson [mailto:sarah at telergy.com]
Sent: 27 February 2000 19:13
To: Music Bar
Cc: Jay Vaughan
Subject: Alien Project track: My alien is GAY!


Here's the page:

http://www.samplelibrary.net/browse.php?num=2&m_id=604&type=12


Every sound on the track is 100% taken from the 'Feed the Aliens' sample. My
first step was to remove the DC offset that caused a click when looping the
file. I did this with a Kyma high pass filter set to 80Hz and 20 poles.
This cleaned it up a treat.

Next step was to create a kick drum. This came from the 'ee' in feed, with
an envelope set to snap the pitch down from about 800Hz to 30Hz very
quickly. This was followed by a low pass filter driven from the same
envelope to cut out unwanted harmonics. This was compressed heavily in
Paris, with a tiny amount of chorus added to fatten it up a bit. Some EQ was
added to bring out the click a bit more.

The closed hat sound is the trailing 's' from Aliens cut up with an
amplitude envelope applied to it. This didn't quite 'crunch' enough, so I
used an interated waveshaper, which is a bit like a distortion unit, but it
has slightly weirder results. After a bit of tweaking, an amazingly
authentic hat popped out. The original sounds a lot more real than it does
on the mp3, incidentally.

The bass sound is taken again from the 'ee' in 'Feed'. It is a short loop of
a few cycles. Two samples are played, one tuned 10 cents up, the other 10
cents down. Both of these were then mixed and fed to a gain stage, which
heavily overdrove a 4 pole LPF. A single ADSR controlled the LPF and the
filter cutoff - a small amount of resonance was used, but not too much. This
was heavily compressed in Paris with a small amount of chorus added.

The cut-up 'rap' sound was created by doing a spectral analysis of the
vocal. This was then played back using an oscillator bank, with the playback
time snapped around using a white noise generator through a sample & hold
triggered by the RM1x. A bit of smoothing was applied to the playback point.
This was compressed in Paris and EQ'd a bit, with a multitap delay added.
The delay time was tweaked until it sounded good rhythmically.

The sliding-up-and-down vocoded sound is actually a modified Kyma factory
patch which uses FFTs in real time to cross the spectrum of a sample (I used
Feed the Aliens") with a recording of dolphin noises. OK, I cheated a bit,
but it sounded so cool I had to keep it. Some multitap delay was added. The
Paris EQ was used to chop off everything below about 500Hz to reduce
clutter.

The slowing down vocal was again done with an oscillator bank, playing the
spectral analysis over and over again slower and slower. Paris settings were
as for the sliding-up-and-down sound.

The wash sound in the intro that comes in quietly was done using granular
synthesis. I intended to do more with it, but it cluttered the mix too much.
Paris was just used to fade it in.

The 'distant' buzzy sound in the breakdown is the vocal set up as a sample,
through a harmonic resonator with a longish decay time. In this case, Paris
was used to heavily mangle the original to make it sound distant. This was
done with a combination of multitap delay, a 10 second plate and an 8khz
high pass EQ setting.

Mastering was done by writing the mix as 24 bit PAF files in Paris, then
reimporting them into a new project and supernormalising them. These files
were then combined into a 24 bit WAV file, dithered down to 16 bits using
noise shaping and encoded to MP3 using bladeenc.

So that was that!
Sarah



------------------------------------------------------------------------
GET A NEXTCARD VISA, in 30 seconds!  Get rates
as low as 0.0% Intro APR and no hidden fees.
Apply NOW!
http://click.egroups.com/1/967/1/_/490048/_/951685844/
------------------------------------------------------------------------




dupswapdrop -- the music-dsp mailing list and website: subscription info,
FAQ, source code archive, list archive, book reviews, dsp links
http://shoko.calarts.edu/musicdsp/




More information about the music-dsp mailing list