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.

Secrets of the IOS Masters @ScottVanZandt and Nanostudio2 (No AUv3 instruments)

McDMcD
edited June 24 in General App Discussion

@ScottVanZandt released a ground-breaking project created with Nanostudio 2 in May of 2019.
On SoundCloud he wrote this about the music track:

Composed entirely on iPad Pro in NanoStudio 2 using custom Obsidian patches created by sampling instruments from the expansion packs of various iOS apps such as Beathawk, iSymphonic, Garageband, Sampletank and Korg Module.

Concept MusicFantasyDarkLightGame MusicFilm Score

Listen to a bit to get the impact of what follows and how it was prepared with weeks of sampling work.

I reached out to Scott for more details about his process and he shared a lot of information and has allowed me to post this thread about his approach.

ME: Can describe the workflow you used to make your NS2 Obsidian samples from iSymphonic and such?

Scott: I spent almost a month sampling all these instruments, not even sure how stable it work in the end, but at 41 tracks in NS2 and the CPU was only at 30% usage, so this method beats the hell out of AUv3, for this style of music that is. Synth apps don’t seem to be too much of a problem loading as Audio Units, but iSymphonic and Beathawk, you’re lucky if you can get 3-4 instances of them running simultaneously before crashing. Having the instruments sampled instead and ready for immediate use is so liberating. And you get so much more control over the sound once it’s in Obsidian.

I record one long audio file by sequencing 4 notes per octave using Korg Gadget Taipei as the sequencer, and record into AudioShare. For example, C1, D#1, F#1, A1, C2, D#2 etc all the way up to A6, or the highest note that will produce sound. I make each note play around 30 seconds, that way you don't have to worry about creating loop points later on which is really hard to get right and will never sound very good in my opinion. Be sure to leave plenty of silence between each note. I like to transfer these long audio files to my iPhone and do the splicing there, throughout the day. It's extremely tedious and time consuming, so I do it when I'm on the bus to and from work, waiting in line at the grocery store, when I need a few minutes to break from work. I have a short attention span, so I spread it out into small chunks. Each file needs to be named like this: ExpressiveViolin C1, ExpressiveViolin D#1, etc. No spacing in the instrument name followed by one space, then the note name and octave.

All of these individual notes once ready, need to go into a folder and it must be a zip file. You can zip it in Audio Share. Once you are in NanoStudio 2, hit the plus sign to create a new Obsidian Track. Open up Obsidian and click Edit. In the top left where it says Type, change the type from Analog to Sample. Tap on VOICE in that same window and change the Polyphony from 4 to 16. Go back to OSC 1 and click Automap Samples. Tap the 3 dots and click import. You can import from Audioshare, or Dropbox in my case. If from Dropbox, just just click the Zip file and it will switch back to the NanoStudio App and say Success. Double click the folder and then double tap on any note in the folder, might as well be the first one on the top. There, all of the notes are automapped. Underneath the tiny keyboard where it lists the note names, click on the first note. Click the wrench icon above the tiny keyboard. Tap Set sample loop, and be sure it's switched to Off. Then, I like to click on PERFORM in the left menu and turn the Release knob up to around 40% so it's not so harsh sounding. Then if you want, you can click MOD/FX and Turn on the Reverb - I like Plate 2. Be sure to click on PATCH in the left menu, click the 3 horizontal lines in the top right (hamburger button), and Save As. Note, I usually end up turning the reverb off at some point and applying reverb to the actual track and not the sample. You get more control options like Dry and Wet. It's just nice having the reverb setup at the patch level for the initial stages of composing so it sounds closer to what it will sound like in the end.

Oh, and if you want 3 velocity layers, which is what I did for my Timpani, you have to do the process 3 times. For OSC 1, 2, and 3. Just use your ears to figure out how many layers the instrument has. Some only have 2, some have several (like pianos), so you should figure out when 3 velocity levels will work best. You can them make adjustments to the velocity levels that trigger the different samples per note. Most of my sustained string, brass, and woodwind samples were just one layer thankfully :)

Oh sure, you're welcome to share my process. Feel free to reword or reorganize how you see fit. I can be a bit scatter-brained, haha. Just recently I've created a few patches using samples I took from Pure Synth Platinum. There are some really nice ones in there. Sometimes when I run out of ideas, it's nice to just keep adding to my Obsidian instrument library for future use. The clarinet, oboe, and brass ensemble from Garageband are really nice sounding to me. I found a really nice solo flute in Music Studio called Flute Vibrato - just added that one yesterday - not sure if that one came from the expansion pack or the original sounds included with the DAW. The french horns I use are from Miroslav - I think they sound better than the ones in iSymphonic. Strings from iSymphonic Corda pack are really good, and so are the strings in Korg Module orchestral dreams.

«1

Comments

  • Brilliant. Thank you and well done to both Scott and Walter Cronkite :)

  • Love this!!!!

  • Impressive.
    Manx thanks, @McD

    Move that to the Knowledge Base and possibly the Wiki?

  • There are probably a dozen here that want this capability but who will do the work involved?

    @coniferprod: can you automate some of these steps with a SynthJacker Update please. Especially the
    30 second sampling enhancement. I keep asking and maybe this will show the use case and save Scott a few
    months of sample chopping on the bus. SynthJacker is so close to save a lot of steps in this workflow with the
    Midi Auto-Sampling of AUv3 apps and external audio (which could be another IOS device for those pesky IAA apps).
    And it's pretty solid chopping the recording into NOTE files with the right NS2 naming schemes and folders for 3 layers.
    It also creates a recording file so you can chop and normalize by hand if you need to have more control and apply
    extra FX to the samples for example.

    It's sad we can't share these massive libraries of instruments but it would not be fair to the developers of the great
    ROM-pler apps.

    NanoStudio could contract for higher quality IAP's. I find the Acoustic Packs to be just OK and not stunning like Scott's
    results here.

    "41 tracks (of sampled instrument) in NS2 and the CPU was only at 30% usage".

    Now someone will ask - Can't NS2 just be made into an AUv3? NO! DAW's are huge. AUv3's are intended to be
    light weight and developers pushed right up to the edges using iSymphonic, BeatHawk, Colossus Pianos, Pure Synth Platinum
    and Fret-Less Apps.

  • @tja said:
    Impressive.
    Manx thanks, @McD

    Move that to the Knowledge Base and possibly the Wiki?

    FYI, it is my understanding that things should be put in the wiki and not the knowledge base (which was intended to be a temporary repository until the wiki was running).

  • It's worth pointing out that this workflow could be hugely simplified with SynthJacker, which can automate much of this process and cut out all of the manual labour involved in creating the samples.

  • Thanks for reaching out and sharing the workflow details.

    @richardyot said:
    It's worth pointing out that this workflow could be hugely simplified with SynthJacker, which can automate much of this process and cut out all of the manual labour involved in creating the samples.

    yep.

  • This is 🔥!! Great read, thanks! I also sampled a lot of instruments but I used ‘sample robot’ on Mac to use them in obisidian. That allowed me to sample a ton of sounds in batches automatically. Its so efficient and simply limitless when u resample your own stuff. However the workloads involved are massive. But worth it in the end, in NS2 at least.

    Here’s a video on the workflow that worked for me for anyone interested

  • @richardyot, you must let Scott know about Synthjacker! Maybe @McD can put you in touch. You can always comment on his SoundCloud page. Pming him here can be done, but not certain if he checks in.
    I posted one is Scott’s incredible pieces on the forum. Little or no notice, naturally. It’s hard not to be angry that such a talent (and genuinely nice guy) does not get recognition in terms of financial compensation, but that is the wonderful world of music.

  • @richardyot said:
    It's worth pointing out that this workflow could be hugely simplified with SynthJacker, which can automate much of this process and cut out all of the manual labour involved in creating the samples.

    True except for the 30 second sampling approach. SynthJacker supports 10 seconds max. I have asked @coniferprod to consider extending the time but he isn't convinced it's needed. Hopefully, this will tip the scales for longer sampling.
    The biggest benefit of long samples is the ability to ignore looping for long sustains in the instruments. Looping generally ruins the illusion of the orchestra sound.

  • edited June 24

    Thanks a lot for the look behind the scenes @McD !
    I really love the work from @ScottVanZandt , I’m always impressed when listening his music.
    He is truly an IOS-Master! 😊👍

    P.S
    Did you start a new interview series Secrets of the IOS Masters @McD ?
    Nice! 😊

  • McDMcD
    edited June 24

    @chandroji said:
    Did you start a new interview series Secrets of the IOS Masters @McD ?

    I can see it as a Series of Threads. Ideally there are questions in the comments and the "Master"
    starts answering to keep things moving along. Scott made this very easy with the explicit details of his process.

    Can you nominate anyone to be asked for their process? I'd be interested in learning more from:

    @brambos for creating IOS Products
    @j_liljedahl coder of AUM and Gestrument
    @_ki for his scripting/programming skills
    @richardyot for his studio practices
    @flo26 playing/recording Guitars
    @Linearlineman using improvisation to create Cubasis projects (he uses keyboard splits, mixes real audio with MID, etc)
    @micheal (I could probably just use his Podcast interview to create a written document)
    @analog_matt The AudioKit D1 coding master
    @SevenSystems The MIDI Master-mind

    Anyone else?

  • Well @McKronkite, you may have a new calling... I think “krankheit” means illness. Well, if music addiction is a sickness we are in business.

  • @LinearLineman said:
    Well @McKronkite, you may have a new calling... I think “krankheit” means illness. Well, if music addiction is a sickness we are in business.

    This is beginning to sound like a job. I have quit working for the man!
    If it's fun I'll follow through.

    I think the format is pretty obvious. "How do you use IOS?" Get them to write the text.
    Cut-n-paste. Take credit for the effort.

    The trick is in the Headline. Think of Tabloid Tricks.

    It's fun. Try it yourself using this or any other headline. Email interviews work best to get the text.

  • Nice investigative work @McD
    +Thank you to Scott for being so open and sharing his techniques.. absolutely Golden!

  • How does one sample out of garageband using his technique?

  • @lukesleepwalker said:
    How does one sample out of garageband using his technique?

    I'll ask @ScottVanZandt to see. I imagine it requires more steps: maybe record in GarageBand and export the
    track for AudioShare slicing.

  • Thanks for sharing!

    This method can also be done in BeatMaker 3 .. the steps listed would be very similar.

    also, does Nanostudio only allow 3 velocity layers or is that just the max he ever created?

  • McDMcD
    edited June 25

    @BitterGums said:
    Thanks for sharing!

    This method can also be done in BeatMaker 3 .. the steps listed would be very similar.

    also, does Nanostudio only allow 3 velocity layers or is that just the max he ever created?

    The Synth in NS2 allows samples to be added to the Oscillator and 3 oscillators are the max.

    Scott's ideas make me think I'm making too many layers when 3 is enough for orchestras to still sound good.
    It save RAM and allows for more Obsidians at the same time.

    I was building my orchestra in in AudioLayer and using 5 layers. I would never get to 20 or more instruments live in AUM for example.

    So, his approach and the coding excellence of NanoStudio 2 is making me re-think my approach. By the time I learn NS2 and have a decent set of instruments the audio tracks will probably ship in an update but if they don't I can still make music like this without audio tracks.

  • Ok. Now i get why i need to use SynthJacker...can’t believe I held off this long to use this way with NS2. Thanks for the thread @McD 👊🏼™️

  • @echoopera said:
    Ok. Now i get why i need to use SynthJacker...can’t believe I held off this long to use this way with NS2. Thanks for the thread @McD 👊🏼™️

    Just be advises that SynthJacker maxes out at 10 seconds so you may have to fiddle with adding Loop points for ambient styles. Scott uses 30 second samples so I'm using his work as proof that SynthJacker needs longer sample options to help us avoid looping and all the extra headaches that adds for 76 waves per instrument. Right now synth jacker has some timing issues if the whole sampling exercise seems to go to far beyond 10 minutes total.

    Anyway, this proof of quality shows the SynthJacker dev that we could benefit from longer samples even if we have to
    run in smaller batches like one layer at a time of 24-30 notes. That would be a total recoding time above 15 minutes.
    If you would buy SynthJacker if it have longer sample times let the developer @coniferprod know in a PM. I'm hoping he's working on making the App work with IAA Apps for input too so we can sample our aging Apps like Animoog. He's been heads down for a while so an update could be imminent.

  • @McD said:

    @richardyot said:
    It's worth pointing out that this workflow could be hugely simplified with SynthJacker, which can automate much of this process and cut out all of the manual labour involved in creating the samples.

    True except for the 30 second sampling approach. SynthJacker supports 10 seconds max. I have asked @coniferprod to consider extending the time but he isn't convinced it's needed. Hopefully, this will tip the scales for longer sampling.
    The biggest benefit of long samples is the ability to ignore looping for long sustains in the instruments. Looping generally ruins the illusion of the orchestra sound.

    I'm working with sampled instruments and I would say that except for drones and atmos, 10 seconds is more than enough.
    How often do you hold a piano note for longer than ten seconds? A guitar? Brass? Any orchestral instrument?

    This piece is great work, Wow. A good example of what can be done when you're ready to spend your time making music.

  • edited June 25

    @McD Thanks for this article ..I was very excited when Scott released it, and it's cool to read more "behind the scenes" info ... Really fantastic track, true masterpiece. I like it in many levels, not just as true piece of art, but also as perfect showcase that you don't need powerful computers and plugins worth of thousands of euro or full studio of hardware to make awesome track. For me Scott is exact opposite to people who are all the time just complaining about missing features, missing apps, missing tools, missing this, missing that, they can write whole stories about why they cannot make music - thank god (or what) for people like Scott who have not been told there is something missing or impossible, so they just make perfect music :smile: Very inspirational person, such people helps me to go through my personal laziness (i'm actually very lazy person:)) and actually do something valuable.

    It's incredible what is possible to do with just iPad, great time to live !

    at 41 tracks in NS2 and the CPU was only at 30% usage

    Yeah this is totally crucial thing. Not everybody can appreciate this, many people are into music styles which needs significantly less tracks in project, but if you need 10ths of tracks, lot of FX - this is something incredible. With latest iPad MINI i feel now virtually limitless. I'm able just add and add and add .. layer after layer layer - and CPU is still below 50% :) Even with internal VA oscillators (which are much more CPU hungry than just plain sampling) you can easy reach 100 tracks - if you use sampling instead of VA oscillators, you simply cannot run out of CPU power in any existing meaningful music style :)

  • @rs2000 said:

    True except for the 30 second sampling approach. SynthJacker supports 10 seconds max.

    When you play string parts or play a piano slowly with the sustain pedal you will often hit the 10 second limit.
    Having a maximum of 30 seconds is for those of us that have the issue. We have to loop the samples or hear the strings
    just stop playing. I agree that the piano note might fade out but I often add strings to the piano so I leave some chords hanging for more than 10 seconds.

    I can just do it manually like Scott does but he spent 60 hours making his samples of the orchestra to make this track alone.
    That might take me 3 months and to be honest I probably won't make the effort. But I would launch SynthJacker and build a few instruments a day if I could just check on it every 15 minutes and make a new file set.

    The other issue with SynthJacker's 10 second limit is timing of the slices. I use 10 seconds and get samples near the end of the run that seem to get sliced in 2-3 second results.

    Arg. It's always something that I want changed. Hopefully SynthJacker gets a longer option.

    You can still use 10 seconds. OK?

  • @McD, at last a meaningful contribution! 🥴😂😘 Do more!

  • edited June 25

    can just do it manually like Scott does but he spent 60 hours making his samples of the orchestra to make this track alone.
    That might take me 3 months and to be honest I probably won't make the effort.

    It is worth that effort. Think about fact that this is re-usable in lot other tracks - Scott now can for next few years make lot tracks with these sampled libraries ;)

    I'm working on my own Obsidian banks created from multisamples of HW analog synths i own - to the date i have appros 120 - 150 patches, created from total amount probably 50-60 different multisamples sampled usually F0/F1/F2/F3...F5 .. and then looped in NS2 audio editor (to have not-clicking loops). Something like 'single cycle' waveforms but they are not single cycle but rather various length from 0.5 to 5-6 secnds, to preserver also "movement" and slight changes analog synths oscillators sound ...

    I spend maybe 2 months of work with sampling / settings perfect loops on that samples - and another month on creating patches from those multisamples - but now i have quite nice library of fat interesting sounding sounds (which is still growing) - lot of inspiration for my future tracks :)

  • I want to concur.

    That means agree.

    Because I guess sometimes I don't....... ;)

    Ironically I opened NS2 for the first time last night and had my mind blown.

    It is everything "other" apps promised to be and weren't.

    AMAZING APP.

    AMAZING INSTRUMENT SOUNDS.

    I look forward to seeing how midi config and audio input with hardware works in it and as far as mix downs versus tracks and projects and what is possible.

    So, need to watch some tutorials and read the.........nah......watch some vids.

    But the sound is off the hook.

    The intuitive nature of the GUI...off the hook.

    This is the best multi app I have seen in years.

    Look forward to bringing my own samples into the app.

    First app that I said this may be more fun and productive than hardware for a break.

  • Agree big time! I’ve been noodling around like a kid in a huge candy store! Love it so much and trying to unlearn all the ways of doing things and
    Just having an open mind to learn the NS2 way!

    So much so deep! Super amazing!

    Can’t wait to sample also. Right now I’m trying to learn more about sampling as I want to sample my aging Korg Trinity as the analog outputs are kaput, I’d have to go via digital out to my old XP PC which is also rusting away😌❤️
    But Synthjacker seems exciting and the way to go!

    @RUST( i )K said:
    I want to concur.

    That means agree.

    Because I guess sometimes I don't....... ;)

    Ironically I opened NS2 for the first time last night and had my mind blown.

    It is everything "other" apps promised to be and weren't.

    AMAZING APP.

    AMAZING INSTRUMENT SOUNDS.

    I look forward to seeing how midi config and audio input with hardware works in it and as far as mix downs versus tracks and projects and what is possible.

    So, need to watch some tutorials and read the.........nah......watch some vids.

    But the sound is off the hook.

    The intuitive nature of the GUI...off the hook.

    This is the best multi app I have seen in years.

    Look forward to bringing my own samples into the app.

    First app that I said this may be more fun and productive than hardware for a break.

  • edited June 25

    during sampling directly into NS2 don't forget the hidden gem in NS2 audio editor - "Actions > Save SEL" ... which allows you save just selected portion of sampled audio as separate file where original file stays still opened in audio editor selected - so you can then move using bottom middle handle to another sampled note and save it again as separate file with exactly same size ...

    this, combined with grid quantisation enabled makes creating slices with same length from long recording super easy

    this is super handy for recording one big take for more notes and then save each note as individual file...

  • @dendy said:
    during sampling directly into NS2 don't forget the hidden gem in NS2 audio editor - "Actions > Save SEL" ... which allows you save just selected portion of sampled audio as separate file where original file stays still opened in audio editor selected - so you can then move using bottom middle handle to another sampled note and save it again as separate file with exactly same size ...

    this, combined with grid quantisation enabled makes creating slices with same length from long recording super easy

    this is super handy for recording one big take for more notes and then save each note as individual file...

    That's a good tip, especially for instruments that can't be sampled by SynthJacker.

Sign In or Register to comment.