posted 19 March 2012 16:19            

Anyway, my question is this: I am looking for a down and dirty way to sync footsteps to video. I've toyed around with using a microphone into a amplitude follower to control the gate and level and a frequency follower to control pitch going into a multisample randomly picking footstep audio files from a folder. I then would perform the footsteps with my mouth to the video. I haven't had very good results.

Has anyone tried this? I'm interested to know how some of the Kyma masters on this forum would handle something like this.

Thanks so much.

-Jeff

posted 20 March 2012 19:29            

mouthFoley.kym

One approach might be to route the Input-->AmplitudeFollower--->Threshold, and put the Threshold into a SoundToGlobalController with !Gate as the GeneratedValue. That way you can make adjustments to the level and threshold and see whether the Gate is getting triggered.

Here's an example (but triggering glass samples instead of footsteps).

posted 30 March 2012 10:48            

The reel time (DAW) way of working is better at achieving some kinds of results for sure. But the real time performance of sound to picture can capture a kind of expressiveness, nuance, and reactiveness of a live performance that is extremely difficult to achieve in 'reel time'. The mind/hand/ear/eye (and mouth?) real time control system is capable of some amazing things.

Not saying that one way of working is superior to the other, because we all switch back and forth between these modes, many times a day, choosing whichever mode best fits the situation. I think it's cool that Jeff is experimenting with new approaches to try to bring the expressiveness of a live performance, the human breath & vocalization, to sound for picture. He is doing more than just lining up static recordings; he is also performing the loudness and pitch in real time in response to the picture.

I hope we can all discuss these kinds of things in more depth in Minnesota (http://kiss2012.symbolicsound.com)

posted 30 March 2012 15:05            

My problem is that i would like to prevent a footstep to be triggered two times in a row (random no repeat). And i'm in the Capytalk learning process right now.

posted 30 March 2012 20:04            

randommultisamplenorepeats.kym

code:

---

| original randomized r |
r := Random new.
original := (1 to: ('bell.aif' sampleFileNamesInSameFolder) size) asOrderedCollection.
randomized := OrderedCollection new.
[original isEmpty] whileFalse: [ | randIndex |
	randIndex := (r next * original size) truncated + 1.
	randomized add: (original remove: (original at: randIndex))].
(((1 bpm: !BPM) countTriggers - 1) of: randomized) - 1

---

A verbal explanation would be that it starts with a collection of your original indexes and an empty collection called randomized. One by one, it fills up randomized by picking elements at random from the original, and REMOVING them from the original so they cannot be picked again.

posted 04 April 2012 11:15            

quote:

---

Originally posted by SSC:

code:

---

| original randomized r |
r := Random new.
original := (1 to: ('bell.aif' sampleFileNamesInSameFolder) size) asOrderedCollection.
randomized := OrderedCollection new.
[original isEmpty] whileFalse: [ | randIndex |
	randIndex := (r next * original size) truncated + 1.
	randomized add: (original remove: (original at: randIndex))].
(((1 bpm: !BPM) countTriggers - 1) of: randomized) - 1

---

---

The sound stops after playing all the sounds. I would like it to do an infinite loop (always restart after reaching the end. (with a new random picking))