k y m a • t w e a k y — the kyma collective || view the current website || February 2015 Archive

/ WebHome / Sounds / Share.DiscussTransientEnhancerDiscussion

Search


Share Section


Sounds
Samples
Timelines
Tools
Microsounds

Home 
Topics 
More... 

All Sections


Products
Order
Company
Community
Share
Learn

Login / Register 
Change password
Forgot password?

Symbolic Sound


Home
Kyma Forum
Eighth Nerve

TWiki Links


twiki.org
This sound can be used to either enhance or diminish the sound of attack transients and their release. Probably most useful on drums, but perhaps elsewhere too.

This sound grew out of some experiments where I was attempting to measure mean energy in a sound by the use of a sum of two different peak trackers, one with a short attack and the other with a longer attack.

But instead of summing them for an energy estimate, why not subtract them? Then you get a record of the size of a transient attack. Subtract a slow responding peak follower from another with instantaneous attack. Keep both at the same release. This can be used as a gain control in the following Gain block. The GainA? parameter controls whether to boost the attack portion of the sound (positive dB gains), or attenuate it (negative dB gains).

This first gain stage is followed by another that looks at the release tail of the sound. Both followers have the same attack characteristics but one has an instantaneous release which is subtracted from the other follower with a longer user settable release time. Now this difference indicates something about the amplitude of the release. The following gain stage uses parameter GainR? in dB to control whether or not the release tail is amplified or attenuated.

I haven't tried this sound on anything but drums, but I will.

-- DavidMcClain - 15 May 2004

Here is another SoundFile? containing the original TransientEnhancer? along with a stereo version that works in dB (log) space. The advantage of working in dB space is that you get the same amount of enhancement regardless of how strong or weak the input signal is. Instead we are comparing the difference between the dB level of peak trackers, or the ratio of their values.

There is an additional parameter K that is used to compensate for the scaling of the dB convertors. These dB convertors produce an output which is the dB/192 of the input signal level. K attempts to scale that back to more useful values. In the default case provided we have K = 16, meaning that the gains are GainA? * dBdiff/12 and GainR? * dBdiff/12.

Using too much K, e.g., 32 produces click artifacts in the output sound.

As I'm writing this note, I am listening to a full mix recording being fed through this enhancer with the default settings, Attack 10 ms, GainA? +10 dB -- which enhances the attacks of sounds, and Release = 300 ms, GainR? = -6 dB, which has the effect of cutting the tails of sounds much like a noise gate.

-- DavidMcClain - 15 May 2004

Hi David,

I was off the board for quite some time. So I couldn´t contribute what I was doing in March 2003 and posted on the old forum ("Attack designer"): http://www.symbolicsound.com/cgi-bin/forumdisplay.cgi?action=displayprivate&number=3&topic=000095

I didn´t look at your sound since my head is somewhere else right now but it seems we had a quite similar approach :-)

Bests, - Mathis -- MathisNitschke - 25 Sep 2004

 
 
© 2003-2014 by the contributing authors. / You are TWikiGuest