How would I get this lovely sound effect... hopefully with synths not samples

Please listen to this tester from an old friend of mine… I’d love to
recreate the sort of… ‘breathy’ effect of the notes, but I’m just not
finding the right fx’s to do it…

Any help or code would be appreciated. Truly.


Do you mean the distant/muted breathy notes in the background in particular?

Yes, that… but I’m also interested in how he gets the strong foreground notes to… change pitch, or detune… not
figured that one out yet either…

Most of the drum/hat/cymbal stuff I’ve got figured from making house but… ohhhhhh, those notes. :slight_smile:


Portamento on the synth voice to get the dip up on the pitch.
The main voice is also being sent to a long reverb and the reverb is sidechained to the dry, main voice, so that the reverb ducks every time the main voice plays.
In the first part of the song, you hear the verbed voice without any dry signal and not the main instrument. So either a mixer setting of, pre-fade fx send etc, to achieve the wet only sound.

At least that’s how I would do it.

Hope that helps.

You lost me at: Portamento.

(So I looked it up and also searched the forum) :slight_smile:

However, the stuff I mention sounds like it was done in a DAW, such as Ableton…

I want to know how to do it in SP, using code, not some knob-twiddling
thingie… so if you have any examples, I’d appreciate a look.


Considering I’ve been using SPi for a sum total of three weeks, I would suggest the following.
In the live loop where you are playing the melodic figure, you also have a duplicate playing that is surrounded in an fx wrapper with the mix set to 1.0. This way the same notes can serve both parts.

pluckMel = (ring 60, 57, 60, 65, 64, 60)

live_loop :melPlayer do
  #let's assume you have a ring of notes called pluckMel
  play pluckMel.tick, release:0.6, amp:0.6
#ghost melody
  with_fx :reverb, mix:1.0, amp:0.2 do
    play pluckMel.tick, attack:0.2, release:0.1
  sleep 0.75

The simulation of sidechain compression i.e. self, is going to be harder and I imagine that will use something like a fadein/fadeout type code. I’ve seen a few examples on the forum that could be adapted. Whilst it is easier in a DAW, I’ve not yet come across a solution implemented in SPi. Personally it would be good if instruments had a S/C listen opt, something like cue but for sidechaining.

The melodic figure is quite repetitive, so easily transcribed if that’s what you want to do.

Chords are ii-I with the bass moving between ii, I, ii, vi.

Portamento is pretty much standard for instruments and synths, especially for trills without retriggering. In other implementations it is called glide etc and, such as TB303, is a characteristic feature.

The code above is rough and will only get you part way there. You can explore the additive tutorial Sam has, where perhaps combine the plucksound with one that also has some noise content, which you should also treat with the hpf, set at about 85/90, so that only the high noise values will poke through. A combination of these might get you closer to your goal.

BTW: my advice was not knob twiddling focused. It was a rough outline of what you would need to achieve in SPi. Having only been here a few weeks, I’ll keep my head down.

Hi Hussein,

‘Knob-twiddling’ was not a reference to you. Please dont think that.

-MY- perception of the sounds my friend produced was that it was done
with a DAW… but then, I have no experience of DAWS to confirm that,
I’m simply some old coot who tinkers with SP, and wants to know more.

It also sounds to me like you know a lot more about music theory and generation,
than I ever will… I’ve been around SP for about 4 years, and it’s obvious you know
more than me, my friend, so I welcome your input, wether you have used SP for
3 weeks, 3 months, or 3 years.

No, please dont keep your head down… thats not what our forum is about,
if you have an opinion, or code that will help others, please present it… it
will be accepted by everyone.

It may well be used as a pivot point for finding an easier way to do things…
people will take it, and make it their own, modifying it, and feeding it back…
but I assure you it will not be criticised.

Let me offer this post to you, of the kind of fun we have round here: :slight_smile:



1 Like

Thank you.

If not sidechain, you can also use simple volume change. A potential extension being placing the sidechain/volume in a separate block, then using cue with a delay to trigger. You’ll have to pass the block, the melodic line and sleep values etc.

Yes, in your post you already have a lot of the controls/slide code, so probably a question of adaptation. Changing the curve from linear to something a bit more exponential.

Good luck with modelling the track.


Ah… so you mean something like this…

set_mixer_control! amp: 1

slider = 0.125

chords= (scale :e3, :minor_pentatonic)
chords = (ring :c4, :e3, :c4, :c2, :c3, :e4, :d4, :e4)

live_loop :siren do
  use_synth_defaults amp: 0.5, cutoff: 100, env_curve: 6
  siren = synth :beep,  note: c, note_slide: slider, sustain: slider
  sleep 0.25
  control siren, note: c+12
  sleep 1

I’m sorry, I’m not good with the techincal words about sounds… I understand
code better, but I’m not really a coder either… or a musician for that matter.

So sometimes it takes me a while to see what people mean… :frowning:


Yeah, a sort of portamento/slide. The curve would need to be a bit sharper. Slider seems to divide the rise time into two equal portions. At least that’s what it sounds like.

It’s at this point where my inexperience and knowledge of Sonic Pi fails, when applied to preexisting production practices in DAWs, synth playing etc. I just haven’t investigated that bit of SPi yet, so don’t know what alterations could be made that would create a smoother model.

One alternative I could suggest is to use a sample that has this sort of character baked in as part of its sound. So you could have both a clean foreground synth and the reverb tail come up behind it as separate samples. Here’s an article that is somewhat similar.

In terms of the other stuff, its going to emerge and adapt over time. So enjoy!
I can recommend Gordon Reid’s Synth Secrets Sound On Sound website in case you haven’t read them. Covers pretty much everything.

Hope some of that helps.

1 Like