Audiobus: Use your music apps together.

What is Audiobus?Audiobus is an award-winning music app for iPhone and iPad which lets you use your other music apps together. Chain effects on your favourite synth, run the output of apps or Audio Units into an app like GarageBand or Loopy, or select a different audio interface output for each app. Route MIDI between apps — drive a synth from a MIDI sequencer, or add an arpeggiator to your MIDI keyboard — or sync with your external MIDI gear. And control your entire setup from a MIDI controller.

Download on the App Store

Audiobus is the app that makes the rest of your setup better.

Favorite way to randomly trigger phrases of music

I've been thinking about Brian Eno's approach to generative ambient music, where (if my understanding is correct) he will basically set a bunch of undulating semi-random sequences of differing lengths in motion and then sort of capture the best bits and crossfade them together. This is how I understand Reflections (expert) was done. I think he actually worked with a programmer to make this happen in the app and probably used that design for the album we bought or listen to on Sp*tify or whatever, but I wonder how one would approach this method in an Audiobus setup. One thing I was thinking was to make long 6, 8, 16 bar samples, map them to keys, and trigger with a midi sequencer that does gentle randomization such as Xynthesizer. One thing I've done in the past is create loops of different lengths in Loopy and just play them until I get sick of it.

Any thoughts?

«1

Comments

  • Sounds like we're headed in the same direction. Q? How are you crossfading again? You've got 32 steps in Xynthesizr and velocity(volume) you can fade in and out. And midi channels on same to route parts to. You're triggering the sound swaps via midi keyboard. Correct?

  • I was planning on using LP-5 instead of Loopy for the additional fx slots per channel.

  • I think Looptunes lets you cross fade it's pretty sleek.

  • edited August 2017

    I tend to use the iPad to record ambient 'jams' as live audio bits, then throw a random selection into Soundforge on the PC for mixing together, though I sometimes use Auria. Soundforge is quicker though.

    Don't use apps generally to do this, as my audio files are pretty long and it'd get a bit messy.

  • @gleandibson said:
    I've been thinking about Brian Eno's approach to generative ambient music, where (if my understanding is correct) he will basically set a bunch of undulating semi-random sequences of differing lengths in motion and then sort of capture the best bits and crossfade them together. This is how I understand Reflections (expert) was done. I think he actually worked with a programmer to make this happen in the app and probably used that design for the album we bought or listen to on Sp*tify or whatever, but I wonder how one would approach this method in an Audiobus setup. One thing I was thinking was to make long 6, 8, 16 bar samples, map them to keys, and trigger with a midi sequencer that does gentle randomization such as Xynthesizer. One thing I've done in the past is create loops of different lengths in Loopy and just play them until I get sick of it.

    Any thoughts?

    Very interested by this question.
    On DrumPerfect Pro, we can load samples/loops up to 20s ( anything, not just drum loops... ). Lot of them, as 16 slots are available for each of the sixteen instruments. Then sequence ( with probability ) the playing of those loops, which will be randomly chosen inside each instrument. There is literally infinite possibilities, and can go on for hours. Audio results and can of course be recorded/exported.

  • @audiblevideo said:
    Sounds like we're headed in the same direction. Q? How are you crossfading again? You've got 32 steps in Xynthesizr and velocity(volume) you can fade in and out. And midi channels on same to route parts to. You're triggering the sound swaps via midi keyboard. Correct?

    I crossfade in "post production" with Reaper for now. Right now I am just using Xynth as a trigger for random notes. I haven't really come up with a good way to trigger whole phrases with it, but this thread is already giving me ideas.

  • @MonzoPro said:
    I tend to use the iPad to record ambient 'jams' as live audio bits, then throw a random selection into Soundforge on the PC for mixing together, though I sometimes use Auria. Soundforge is quicker though.

    Don't use apps generally to do this, as my audio files are pretty long and it'd get a bit messy.

    This is very close to how I've done things. One thing I've tried is creating stems to load back into AUM then "playing" pans and crossfaders with controllers such as tb midi stuff. With mixed results as I often lose interest in the track before getting that all customized.

  • @Gilbert said:

    @gleandibson said:
    I've been thinking about Brian Eno's approach to generative ambient music, where (if my understanding is correct) he will basically set a bunch of undulating semi-random sequences of differing lengths in motion and then sort of capture the best bits and crossfade them together. This is how I understand Reflections (expert) was done. I think he actually worked with a programmer to make this happen in the app and probably used that design for the album we bought or listen to on Sp*tify or whatever, but I wonder how one would approach this method in an Audiobus setup. One thing I was thinking was to make long 6, 8, 16 bar samples, map them to keys, and trigger with a midi sequencer that does gentle randomization such as Xynthesizer. One thing I've done in the past is create loops of different lengths in Loopy and just play them until I get sick of it.

    Any thoughts?

    Very interested by this question.
    On DrumPerfect Pro, we can load samples/loops up to 20s ( anything, not just drum loops... ). Lot of them, as 16 slots are available for each of the sixteen instruments. Then sequence ( with probability ) the playing of those loops, which will be randomly chosen inside each instrument. There is literally infinite possibilities, and can go on for hours. Audio results and can of course be recorded/exported.

    Thanks for the tip - will definitely look into that.

  • @gleandibson said:

    @MonzoPro said:
    I tend to use the iPad to record ambient 'jams' as live audio bits, then throw a random selection into Soundforge on the PC for mixing together, though I sometimes use Auria. Soundforge is quicker though.

    Don't use apps generally to do this, as my audio files are pretty long and it'd get a bit messy.

    This is very close to how I've done things. One thing I've tried is creating stems to load back into AUM then "playing" pans and crossfaders with controllers such as tb midi stuff. With mixed results as I often lose interest in the track before getting that all customized.

    iDensity is pretty good, though it does degrade the sound a bit. Borderlands too (though it's needing an update), has an excellent visual way of working with audio files.

  • Worth mentioning Wotja of course, which the original app it started out as Eno used. It's a bit frustrating gui wise (but the developers did implement a dark mode after a nudge from me so they are constantly listening to feedback).
    But there's a great engine in there. You can import samples too but the workflow is a bit confusing at first. Hoping for Ableton Link at some point.

  • Think the classic Eno method is to use loops of differing lengths with silence inserted. So you might have 8 tracks with 8 seconds of audio on each but each track loop is anywhere between 20 and 60 seconds in length so that they always layer differently. If you turn off 'track quantize' in Loopy, you can definitely make this happen—provided you record enough silent space in each loop.

  • @MonzoPro said:

    @gleandibson said:

    @MonzoPro said:
    I tend to use the iPad to record ambient 'jams' as live audio bits, then throw a random selection into Soundforge on the PC for mixing together, though I sometimes use Auria. Soundforge is quicker though.

    Don't use apps generally to do this, as my audio files are pretty long and it'd get a bit messy.

    This is very close to how I've done things. One thing I've tried is creating stems to load back into AUM then "playing" pans and crossfaders with controllers such as tb midi stuff. With mixed results as I often lose interest in the track before getting that all customized.

    iDensity is pretty good, though it does degrade the sound a bit. Borderlands too (though it's needing an update), has an excellent visual way of working with audio files.

    I love me some iDensity. One of my favorite things is to load up separate AU tracks in AUM and trigger notes with a midi generator. When you said the audio degrades are you just talking about the granular effect or does it actually down-sample the audio? Is Borderlands worth getting if I already have & like iDensity?

  • @syrupcore said:
    Think the classic Eno method is to use loops of differing lengths with silence inserted. So you might have 8 tracks with 8 seconds of audio on each but each track loop is anywhere between 20 and 60 seconds in length so that they always layer differently. If you turn off 'track quantize' in Loopy, you can definitely make this happen—provided you record enough silent space in each loop.

    This is what got me thinking about creating this thread. I want to be able to trigger midi phrases, more than just samples. I was thinking maybe it could be done by toggling on/off clips in Modstep with midi if that's possible. I don't know how flexible Modstep is with lengths of sequences, though. Like, messing with time signatures. I guess what I want to do basically comes down to is creating really slow polyrhythms? Might be cool to trigger long clips with something like Concentric Rhythm at a really slow BPM.

  • @gleandibson said:

    @syrupcore said:
    Think the classic Eno method is to use loops of differing lengths with silence inserted. So you might have 8 tracks with 8 seconds of audio on each but each track loop is anywhere between 20 and 60 seconds in length so that they always layer differently. If you turn off 'track quantize' in Loopy, you can definitely make this happen—provided you record enough silent space in each loop.

    This is what got me thinking about creating this thread. I want to be able to trigger midi phrases, more than just samples. I was thinking maybe it could be done by toggling on/off clips in Modstep with midi if that's possible. I don't know how flexible Modstep is with lengths of sequences, though. Like, messing with time signatures. I guess what I want to do basically comes down to is creating really slow polyrhythms? Might be cool to trigger long clips with something like Concentric Rhythm at a really slow BPM.

    The cool thing about Modstep is that you have a lot of control of the loop length and you can use "follow actions" to automate the movement through a "song" or "ambient track".

  • edited August 2017

    @gleandibson said:

    @syrupcore said:
    Think the classic Eno method is to use loops of differing lengths with silence inserted. So you might have 8 tracks with 8 seconds of audio on each but each track loop is anywhere between 20 and 60 seconds in length so that they always layer differently. If you turn off 'track quantize' in Loopy, you can definitely make this happen—provided you record enough silent space in each loop.

    This is what got me thinking about creating this thread. I want to be able to trigger midi phrases, more than just samples. I was thinking maybe it could be done by toggling on/off clips in Modstep with midi if that's possible. I don't know how flexible Modstep is with lengths of sequences, though. Like, messing with time signatures. I guess what I want to do basically comes down to is creating really slow polyrhythms? Might be cool to trigger long clips with something like Concentric Rhythm at a really slow BPM.

    You could try triggering modstep clips with Audiobus 3 midfilters with the random filter and art kerns probablility gate etc Not tried this but could be interesting.

  • ^ good ideas above.

    @gleandibson said:

    @syrupcore said:
    Think the classic Eno method is to use loops of differing lengths with silence inserted. So you might have 8 tracks with 8 seconds of audio on each but each track loop is anywhere between 20 and 60 seconds in length so that they always layer differently. If you turn off 'track quantize' in Loopy, you can definitely make this happen—provided you record enough silent space in each loop.

    This is what got me thinking about creating this thread. I want to be able to trigger midi phrases, more than just samples. I was thinking maybe it could be done by toggling on/off clips in Modstep with midi if that's possible. I don't know how flexible Modstep is with lengths of sequences, though. Like, messing with time signatures. I guess what I want to do basically comes down to is creating really slow polyrhythms? Might be cool to trigger long clips with something like Concentric Rhythm at a really slow BPM.

    Can you trigger a modstep clip via MIDI? If so, you could use something like midiSTEPs to create differing length trigger patterns that run at different speeds, etc. When the next version of midiSequencer (Quantum) is released, it will open this whole thing right up.

    You could also pad the modstep patterns with rests to adjust the timing.

  • edited August 2017

    @syrupcore said:
    Think the classic Eno method is to use loops of differing lengths with silence inserted. So you might have 8 tracks with 8 seconds of audio on each but each track loop is anywhere between 20 and 60 seconds in length so that they always layer differently. If you turn off 'track quantize' in Loopy, you can definitely make this happen—provided you record enough silent space in each loop.

    So all you need is a midi triggerable/retriggerable sample player a long enough loop and mute bottoms or faders?

    Launchpad app with samples imported?

  • Or just use Eno's Scape app ;)

  • @syrupcore said:
    ^ good ideas above.

    @gleandibson said:

    @syrupcore said:
    Think the classic Eno method is to use loops of differing lengths with silence inserted. So you might have 8 tracks with 8 seconds of audio on each but each track loop is anywhere between 20 and 60 seconds in length so that they always layer differently. If you turn off 'track quantize' in Loopy, you can definitely make this happen—provided you record enough silent space in each loop.

    This is what got me thinking about creating this thread. I want to be able to trigger midi phrases, more than just samples. I was thinking maybe it could be done by toggling on/off clips in Modstep with midi if that's possible. I don't know how flexible Modstep is with lengths of sequences, though. Like, messing with time signatures. I guess what I want to do basically comes down to is creating really slow polyrhythms? Might be cool to trigger long clips with something like Concentric Rhythm at a really slow BPM.

    Can you trigger a modstep clip via MIDI? If so, you could use something like midiSTEPs to create differing length trigger patterns that run at different speeds, etc. When the next version of midiSequencer (Quantum) is released, it will open this whole thing right up.

    You could also pad the modstep patterns with rests to adjust the timing.

    Yes you can midi learn the clips in modstep. Another fun simple random sequencer is Brambos apps which you can feed through AB3 probablility filters etc. Lots of fun. I'm looking forward to the drum sequencer when it arrives :)

    And Quantum, when is it coming?

  • @gleandibson said:

    @MonzoPro said:

    @gleandibson said:

    @MonzoPro said:
    I tend to use the iPad to record ambient 'jams' as live audio bits, then throw a random selection into Soundforge on the PC for mixing together, though I sometimes use Auria. Soundforge is quicker though.

    Don't use apps generally to do this, as my audio files are pretty long and it'd get a bit messy.

    This is very close to how I've done things. One thing I've tried is creating stems to load back into AUM then "playing" pans and crossfaders with controllers such as tb midi stuff. With mixed results as I often lose interest in the track before getting that all customized.

    iDensity is pretty good, though it does degrade the sound a bit. Borderlands too (though it's needing an update), has an excellent visual way of working with audio files.

    I love me some iDensity. One of my favorite things is to load up separate AU tracks in AUM and trigger notes with a midi generator. When you said the audio degrades are you just talking about the granular effect or does it actually down-sample the audio? Is Borderlands worth getting if I already have & like iDensity?

    Borderlands hasn't been updated for over two years, not sure how much it is but I'll try and test it tomorrow to see if it's still working - unique app though.

    I think it's the granular thing with iDensity - I struggle to get a clean sound from it - Samplr does that a bit too.

    BM3 is a favourite at the moment though, the sampler in that is superb so I'll be experimenting with that one over the coming months.

  • I think the main 'Eno' thing to consider here is that the composition part of this is absolutely not random. He sets up specific systems to generate the final piece. It can sound like these things are happening at random (and in the end, they are as they're no longer under the composer's control) but the randomness isn't an input to the system in any sense. It's an output.

    Not to say that he isn't know for experimenting with random generators or indeed using the game of life (ala xynthesizer) as an input to music generating systems. But the basic idea of all of the old ambient works (and some of the newer ones, best I can tell) is to set a bunch of composed, specific things in motion in a composed, specific way and let the pieces fall.

    And, of course, do whatever you want to do! Just commenting specifically on the best of my understanding of how Lé'Eno does it.

  • @Carnbot said:
    And Quantum, when is it coming?

    Soonish! The (hopefully) final beta is out in the wild. Quantum has a ton of features that would support the idea of randomly triggering a midi destination as well as follow actions and an amazing set of clock divisions. The clock divisions alone would sort of take care of needing to pad the source with silence. For sure, it's polymetric heaven.

  • The Music for Airports 'Score' shows the cyclical nature. None of the inputs are random. They're mechanical.

    image

  • @MonzoPro said:

    I think it's the granular thing with iDensity - I struggle to get a clean sound from it - Samplr does that a bit too.

    BM3 is a favourite at the moment though, the sampler in that is superb so I'll be experimenting with that one over the coming months.

    .
    Yeah, I try to use the rounded grain shape and put a ton of reverb on it for that nice shimmer effect. That and percussion, but yeah for the most part it's a pretty harsh sound and frankly a tad passé-sounding at its harshest imho.

  • @syrupcore said:
    The Music for Airports 'Score' shows the cyclical nature. None of the inputs are random. They're mechanical.

    image

    coooool

  • sounds - check
    faders - check
    fx - check

  • @audiblevideo said:

    sounds - check
    faders - check
    fx - check

    Similar to mine, except I've got the 144.

    Bought it for £125 second-hand nearly 30 years ago. It was in mint condition and I made hours and hours of music with it. My favourite trick was to turn the tape over when recording guitar lines for backwards effects.

  • Looking for an extremely slow generative sequencer to trigger ambient samples? Toa. One note every 32 bars.

  • Sorry it is Zoa actually - and it's one note every 16 bars. Several types of generative techniques can be applied.

  • @suitcase4 said:
    Sorry it is Zoa actually - and it's one note every 16 bars. Several types of generative techniques can be applied.

    Zoa is really cool. You can always use notes on a piano roll to trigger something. Like atom2 inside AUM. Then just put the notes wherever you want to trigger whatever app you’re using.

Sign In or Register to comment.