One Sample Dare #43

[ablaut]

Tuonela sent me a sample of a marching band, which I mixed with the voice of a Japanese girl. Now the challenge is for you to make some interesting music out of this. Here's the sample:

https://drive.google.com/open?id=0Bygnh-iFjp7ZUjNvRU9VYkpBSGM

If you decide to accept this challenge, you must follow the rules below:

==================================================

The idea of the contest is to create ambient track using only one sample.

Rules are:

1. Additional samples, soundfonts, synths, samplers are NOT permitted (except instruments which will use only given contest sample loaded). Any other processing of contest sample is permitted.

2. You can use as many copies of original sample as you want, each processed with respect to Rule 1.

3. Time for submissions is 16 days, counted to midnight, UTC, of 16th day.

4. Submissions should be uploaded to place available for all AO members. Links should be posted in thread dedicated to given round. Please, do not use hosting with countdown, CAPTCHA and other obstacles.

5. Sample is provided under CC0 (public domain) license.

6. Winner of contest is chosen by votes of AO members.

7. Winner of contest chooses sample for next contest.

8. Participants SHOULD disclose the formula for winning track for the benefit of community.

==================================================

Good luck!

[ian]

samplers- de la mancha scylla, fruity granulizer, il harmor
fx - fruity parametric eq 2, izotope trash 2, il maximus, nusofting anything mod, valhalla freqecho, valhalla ubermod, valhallaroom, fruity convolver, hy-tp, soundhack +pitchdelay, soundhack +bubbler, fruity phaser, fruity stereo shaper, de la mancha truc2.1, il vocodex, valhalla spacemodulator, fruity flanger
software - fl studio 12, paulstretch

i divided this project up into seven stages with seven different project files, mainy due to cpu constraints and resampling stuff.

in stage one i only had a few instances of the samples but they all had a lot of fx on them. the first thing i did was load 3 copies of the original sample into scylla, with which i did granular stuff. the fx chain on this was trash 2 (multiband distortion, convolution) > spacemodulator (heavy chorusing) > eq (removed noise caused by the granular synthesis) > stereo shaper (panning automation) > convolver (reverb to round it off). the next thing i made was a drone in harmor; i started the sample in about the middle when the marching band was the loudest and froze it, then engaged the harmonizer, harmonic prism, phaser, chorus, tremolo, and mild distortion, with detune division of 2 and 5 voices of unison. the fx chain on this was eq (some notches for texture) > truc2.1 (additional texture and grit) > anythingmod (fast volume gate) > vocodex (frequency distribution) > pitchdelay (small delay time and octave jump for grit) > eq (aggressive notching) > hy-tp (slow tremolo) > convolver (dark reverb). next was the original sample of the japanese girl, which i cut up and timestretched to remove the bits of silence. i put +pitchdelay and +bubbler on it to give it a weird and indistinguishable metallic sound then further bastardized it with an instance of anything mod, which is doing mild ring modulation. i cut out the lows and highs with my eq and then rounded it off with more convolution reverb. the last thing i made here was a weird pluckish sort of texture by resampling the japanese girl in harmor. this had heavy harmonic prism, harmonizer, tremolo, pluck, and 9 voices of unison with 5 voices of chorus. i started the fx chain with the eq because i wanted to take out the highs and lows. next up was pitchdelay, followed by freqecho (both shifted the pitch up aggressively), phaser, and ubermod (as a chorus effect).

with this i ended up with a recording that was about 3 minutes long. i took this and timestretched it to 4 minutes in paulstretch, then i began the second project file. i took the 4 minute recording and reversed it, then layered it over top of another copy of itself that was playing forward... both had no fx on them. next, i made 2 drones. the first one was really low and i created it by resampling the japanese girl in harmor with 1/9 detune, heavy phasing, slight prism, and heavy harmonizer. i put a bandpass filter on this and then added 2 phasers to give it a bunch of motion. the second drone was made similarly to the first but i used anything mod to give it some insane ring modulation and also some convolution. following this, i timestretched the recording of the band and the girl to play for 4 minutes and layered them in with the drones. the last thing i did in this project file was automate a few notch filters on the master to free up headroom and add a little more texture.

in project file 3, i took the samples i had so far and chopped and layered them, eqing both. i opened one of the samples in fruity granulizer and messed with grain spacing and randomness, then played it for the duration of the project. i created two more drones out of copies of the audio clip i had at the end of project file 2. these were made much the same way the ones in the previous project file were made, although one had some heavy resonant peaks on its filters and another had some effect-style convolution. i also had another instance of scylla playing under all this, which again had the sample loaded 3 times and abused the synth's granular functions. i put two more notches on the master and automated those, then bounced the audio out again.

i took this and timestretched it twice; one sample ended up at 1 minute and 20 seconds, the other at 5 minutes. i chopped and layered these and previous samples together. in project file 4 i made heavy use of anything mod. on the master here i created two more automated notch filters, as well as an instance of +pitchdelay which i automated the mix level of.

i timestretched the resultant audio in paulstretch and ended up with a 7 minute clip. i put this in project file 5 and chopped it up, again layering it with chopped samples of previous stages of the recording. almost none of these had a lot of fx on them but nearly all had something being automated. freqecho saw heavy use here, as did hy-tp and the convolver. this project was the only time i used the flanger and one of 2 times i used vocodex. i put all the chopped copies of the samples in groups of 3 and linked each group up to a bus with a multitude of fx (primarily plugins that cause phasing and similar subtractive stuff). i made 2 more drones, only slightly varying the process i used to make them. the lower one had distortion with trash 2 and the higher one had some reverb. on the master i gradually automated a high-pass filter from 30hz to 180hz over the course of the 7 minutes.

in stage six i introduced the vocal drone that can be heard prominently throughout the piece. all that is is the japanese girl sample, but with the silence cut out and valhallaroom (stretched to about 7 minutes and 20 seconds in paulstretch, of course). in stage 6 i took the audio i had in phase 5 and layered a forward and reverse-playing copy on top of each other. each had a volume lfo, but the lfos oscsillated between the positive and negative period of one another so one was nearly silent whilst the other was at full volume and vice-versa.

in the seventh and final phase of the project i again layered a normal and reversed copy of the audio on top of each other. the volume of the reversed clip starts at 0 and the volume of the normal clip begins at 100; as the volume of the reversed clip increases, that of the forward clip decreases. i also messed with the stereo field here, making it so that the samples grew more mono as they became quieter. i timestretched the recording from phase 1 to conform with the length of the 7:20 samples and layered that in as well. it has an instance of anything mod on it which gets its mix level automated. i sprinkled some of the previous drones into the project to add more texture. i did some final eqing, gave it some mild distortion on the mids/highs, and compressed the highs and lows a fair bit. in maximus i adjusted the stereo field so it was slightly more mono than stereo. i rode the gain out so the audio's mainly at a consistent volume but rarely let my transients hit 0 db as i wanted it to have some breathing room. i threw in a little fade out and then i was done.

[Ahornberg]

Funeral March



I took a bit of the marching band and another from the speech and used them unchanged several times.
A short cut of the band is paulstretched 177 times and processed through Kaleidoscope.
I added some reverb and delay in the final stage.

[Tuonela]

Explanation to follow... :timeout:

[sevenism]

Screenshot 2016-08-19 11.33.35.jpg

i've done a picture of the fx chain above. the podfarm fx are:


tone 1: brit amp and cab, lux spring reverb, bubble echo, echo platter, ping pong delay, sub octaves, reverse delay, stereo square chorus, analog square chorus, sine chorus
tone 2: fuzz pi, analog square chorus, synth string, random s&h, tape echo, tube echo, ping pong delay, stereo delay
volumes in podfarm adjusted during recording

so i just slowed down the sample in live8 to 7.5 minutes long and played thru all that fx. i played with the tone volumes in podfarm while recording. some levelling and normalizing afterwards.

[iainx]

My first OSD entry. (Note the chucka-chucka sound and the high pitched squeak are the sounds of the tape decks picked up by the iPhone as it recorded. There's also some random noises of me moving my feet as I played it. If these extra noises disqualify me, that's cool, I had fun making it.)

First I ran it through Paulstretch twice to get some tonal drones. I used Absynth to play these drones with some filters and effects, and then I played them through a few guitar effects: couple of reverbs, a delay and a vibrato. I recorded 4 drones (2xC & 2xE with different filters/waveshappers) to a cassette for about 60seconds. then I cut a 10 second loop out of the tape. Labelled it Drones.

I then cut some of the drums out and ran it through paulstretch with some extreme filtering, put that sample into Absynth, with some filters/effects and recorded that at two different pitches through the guitar effects onto tracks 1 and 2 of a second cassette. Labelled it Space Background.

I then loaded the whole of the sample and a few other filtered versions of it into Borderlands, and played that through delays/reverbs/vibrato for a while to create the wind and crackles, and recorded it as a loop on a Loopstation.

Finally I loaded the vocal part of the sample into Borderlands, and had it playing through the effect chains, controlled by a volume pedal.

Drones tape went into the main 4track, space background went into the second 4 track which was routed into the first. The pedalboard was routed into the main 4 track as well.

Then I played it all through a 1/4W guitar amp and recorded it on my iPhone as I rode the faders/volume pedal/tweaked Borderlands.

[synkrotron]

removed...

[MetaDronos]

"Throlrirberrale"

Synth:
M-RGT Ambitron

Effects:
Bucephal
Chopper
Gigaverb
Kammerl Pitcher
Kerovee
REFLEX PRO2
Skidder
ValhallaRoom
WOK PITCHDELAY

Master channel:
SlickHDR
A1StereoControl
PSP MasterComp
PSP Xenon

[Tuonela]

I did another one... :blush:



The first section is the sample 'improvised' on using a demo copy of The Mangle, treated with delay, reverb and pitch/time variations (this includes the bird noises, which evolved from the spoken section). The last bit is a simpler treatment of the band, slightly stretched and treated with echo and reverb. And birdies...

[davedorgan]

Going Under The Knife



8 Tracks in this.

Track 1: Drones - took two sections of the sample - one where the rest of the marching band joins the percussion, and 2 was part of the poem. In Logic Pro I matched the levels and mixed the two track down on top of one another. In Audacity I - Paul Stretched x 50, then increased tempo by 400%, then used Noise Reduction, then Paul Stretched by 20, and then increased Temp by 100% (now just over 8 minutes). i took that track, duplicated it and reversed it. I mixed down the forward and reversed tracks together and brought it back into Logic Pro. FX were ChannelEQ, Stereo Tool to move it more to the middle, and Eventide UltraChannel.
Track 2: Drones - took the Track one from Audacity and ran it through Ambient_0.3. FX were UltraChannel and Channel EQ.
Track 3: Beep sound at beginning and end. Can't remember how I made this. FX were Space Designer Reverb.
Track 4, 5 and 6: - Choral phrases - Took a phrase of two one syllable words from the poem. In Audacity, Paul Stretched it by 5. In LogicPro, put it in three tracks. FX: PShift, ChannelEQ, Echo and Valhalla Room.
Tracks 6 and 7: Put the poem in Iris 2, track 6 is the rim shot like scratchy popping noise (Space Designer and Channel EQ), Track 7 is the pad sounds (Space Designer and PanCake 2).
Applied just a touch of compression to the mix with Logic Pro Studio FET Compressor - 1.4 ratio and 2db on the Limiter.

[Maharg]

re: Going Under The Knife

Great submission. Good to see another Logic Pro and Audacity user here.

[windspace]

Started off by extracting two short audio clips from the marching band section of the original sample. Also extracted a brief clip from the vocal section.

Track 1: Used the shorter of the two band clips. Loaded it into Sonic Texturizer, started recording and added one voice every second for the first 30 seconds, then continued recording the drone for a bit. Pasted this recording into Reaper, copied and reversed it, and cross faded these two clips. Then copied that and pasted in to extend the length.
Added Fx: Eventide H910 Harmonizer, Smartelectronix Ambience, ReaEQ. Added automation lanes for Smartelectronix Ambience Wet Gain and Diffusion. Used ResEQ to roll off boominess with a Low Shelf.

Track 2: Loaded the longer band clip into Sonic Texturizer and added voices until there were 51 then started the recording. Captured around a minute of that drone. Pasted into Reaper to bridge over the area in the middle of project where Track 1 faded in/out. Added Fx: Ohmforce Frohmage with automation to increase wet level to peak around middle of clip then decrease. ReaEQ Low Shelf to remove boomies.

Track 3: Added Fx receive track. Recieves Track 1. Contains NI Replika XT and Vahalla Shimmer.

Track 4: Short clip of voice, Paulstretched (~x3) three times, trimming audio each time until I had a nice "uhhh" pad. Imported into Iris 2 and applied spectral editing and a LP filter. Added envelopes in Iris for amplitude and LP filter cutoff. Duplicated two more times in Iris and set the root note different for each sample so Iris 2 would play a chord for each MIDI note. Drew MIDI track to play long slow notes using harmony from marching band riff. Added Fx: Frohmage and Shimmer.

Track 5: Cut vocal section from original sample and pasted into Reaper. Reversed the clip and duplicated so that same clip is heard in intro & outro. Added Fx: SoundHack ++bubbler, H910 Harmonizer, u-he MFM2, Smartelectronix Ambience, and another Low Shelf using ReaEQ.
Automated track volume, ++bubbler Density, Ambience Wet Level for intro. The automation lanes were applied in reverse at outro.

Master track Fx: Eventide Blackhole, u-he Satin, and and yat another instance of ReaEQ for a Low Shelf and a shallow Notch at about 175 Hz.

[Ambire Seiche]

track 1 - short sample of horn at 1:47 mark hyperstretched 577x with tonal and compression. added tal reverb iii mono air plate


track 2 - speech portion of sample run through fscape seek and enjoy w/ min chunk length turned up, then lowered volume through amplify in audacity, added bouncing ball delay at default settings, used some noise reduction and amplified back to -0.3. added epic verb,vocal plate small, tds vor slick eq, frequency magnet(part of spectral suite)


fx mixer - sc4 mono, mhorse - terry's house and masquarade - karakter

[Maharg]

In my current new phase of keeping things simple I chose to use the Japanese poem(?) for sound source.
I chose a tiny sample and slowed it down by 50% then paulstretched it 50 times. Added some reverb, looped this 6 times.
I used this for the first 3 tracks which I introduced into the mix in a step formation.
The next 3 tracks were made from the first track. Slowed that down 50% and added more reverb.
I also looped this 6 times and added it into the mix in a step formation.
The last 2 tracks (7&8)use the entire poem. The whole poem was slowed down 200% with a big reverb and repeated twice.
The last track (8) is the same processed poem except with more reverb, and a couple of spreader filters were used to finish it off. Just played the once.
Lots of fine adjustments in the mixing and mastering.

Best listened to with headphones...of course

[arachnaut]

OSD43 - Kwaitnuf - RiverSong



I cut the original sample into two parts - one with the marching band playing the Colonel Boogey March and the other with the female voice.
I resampled these to 48kHz 24-bit and performed some noise reduction in Sound Forge.
Then I put those two samples in Reaktor Travelizer and made a MIDI file to tune the resonators based on the March theme.
I played with Travelizer a bit and recorded a 4 minute track based on the two samples.


Then I edited that track into two samples for further processing.


One sample was 2 refrains of the female voice 'singing' the theme, the other was 2 refrains of the marching band doing the same and at the same length.
These were in different pitches because of the resonator tuning settings, so I used Pitchmap to assign the pitches to the same tonality in these two samples.


I placed the band refrain in a Sonar track and replicated it for about 15 minutes to give me a lot of time to play around.
Then I placed the voice refrain in another track, lined them up with the first, and replicated it to the same length.


For the band track I placed Kore as an effect processor and loaded the Deep Freq 'Xtal Globes' preset. I played around with the controls while it played out.
For the female track I placed Reaktor Molekular as an effect processor and used the 'Pretext' snapshot, playing around with the morphs while it played.


Then I mixed the two effects tracks into the Master using Adaptiverb with the 'Bionic Raytracing A HCF' preset and mixed the track levels and final effects.


While doing this, I think I just about went crazy listening to that River Kwai theme over and over, so I thought some of that should persist in the final version.


I hoped to evoke a river or stream with some fairy singing and then a crazy fiddler came in and wouldn't go away.


Denis Goekdag's stuff features prominently here - Deep Freq, Pitchmap, Molekular and Adaptiverb.