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.

SynthJacker

2456711

Comments

  • AU support ist really great. Now it's time to develop an sophisticated auto loop algorithm (something like the amazing Penrose machine from Redmatica's AutoSampler). It's very hard DSP stuff, I know ;-)

  • V0.4.0 is on the App Store now, with options to post-process the samples (found in the settings panel) - trim silence and normalize.

  • McDMcD
    edited April 2019

    EDITED: Due to my misunderstanding of the app, I didn't know how to use the Noise Floor setting to have the samples trimmed properly.

  • @McD said:
    I tried the Trim Silence with several "AU Hosted" Apps and it's not working for me. I'm assuming it will make the recording and then chop the samples into the labeled files and they will be trimmed if I select Start or Both. Am I missing a "Post Processing" step in my workflow?

    I tried Ravenscroft 275 and Colossus Piano and they have close to 1 second of silence before the samples leaves the zero line.

    Did you adjust the noise floor setting? It determines the threshold below which is considered "silence". This may need to be different depending on whether you are sampling straight from the AU, or using an external audio interface. And also, depending on the instrument.

    There's no other Auto-sampler for IOS so I'm hopeful this one keeps getting better.

    Well isn't it? :wink:

    Have you been able to construct a "hands-off" workflow with MainStage or SampleRobot? Don't they also require a bit of post-processing? I'm just trying to gauge how automatic an auto sampler needs to be, or what is realistic.

  • @coniferprod said:
    Did you adjust the noise floor setting?

    No. I left it at the default.

    Thanks for the clue. I will find the right setting to fix my problem. Awesome.
    This explains why the Bass App sampling worked since I just picked a single velocity of
    50 for the test and the signal was Percussive and much louder than the pianos.

    For the pianos I was eager to capture a lot of layers and of course it starts with the smallest layer and that probably established a pattern for chopping then entire recording.
    I'll make some run with just velocity 20 and find the noise floor setting that applies the best trim results. I'll report back on the status.

  • @coniferprod said:
    Have you been able to construct a "hands-off" workflow with MainStage or SampleRobot?

    I had it on my "to do" list for months. I just didn't relish the iPad to Mac inter issues. So, I jumped on the first IOS alternative to solve the problem.

    Don't they also require a bit of post-processing? I'm just trying to gauge how automatic an auto sampler needs to be, or what is realistic.

    Maybe someone can comment. @Lithalean makes a lot of sample sets for audio layer and is well versed in most of the tools and options.

    I'm hopeful the "noise floor" setting in SynthJacker makes this a 5 star app for me.

  • TESTING STATUS: I've tried multiple "Noise Floor" settings and am getting no samples except for or two showing peaks of -15 and -5 dB. Is there a suggested setting? I think the default was set to -51 dB before I changed it.

    I have tried -60, -50, -40, -20, -10, 0 dB. I am testing with velocities that never failed to produce results before.

    I am excited to see that sampling intervals of 12 steps (an octave) have been added. I have been using the octave so maybe that's a new variable and I should try some 3 steps runs.

    So, still digging.

  • Moving the interval back to 3 steps has given me more samples visible in AudioLayer.
    Still seeking the right "Noise Floor" settings to make the "Start" trim work.

  • I think I just had a useful run using velocities of 30,80, 127 and "Start" trim and Noise Floor -54dB. Getting late so I'll pick this up tomorrow. We moved the needle from 3 stars to 4.

    Once I get this dialed in I want to test the "octave" 12 step setting for some generic synth sounds cloning.

    Thanks for shipping me the clue about Noise Floor. I think -51 dB or whatever the default was need to be shifted down 2-3 dB to make the trim work. I hope I'm not celebrating too early.

  • @coniferprod: Yes! -54 dB Noise Floor makes the Start Trim work on my delicate piano Apps.
    I tested a run with Velocity = 20 and it imported right into AudioLayer with no additional editing needed.

    Now I need to do some multi-layered runs and see if the trim continues to work and save
    the hand editing. I expect I could also do multiple runs with various velocities and keep adding each as an additional layer.

    So nice to have a pure IOS Auto-sampler for AU Apps. I'm eager to see what you add next.
    I think adding AudioBus support where we could create Tracks with AudioBus that feed into your App as the target for sampling. This opens the inputs to a wealth of aging products that many would like to clone for use in NS2 or in the other DAW's using AudioLayer.

    I should test the SFZ file you create in Auria Pro when I think about it. Lyra is a great sampling engine with rock solid disk streaming.

  • McDMcD
    edited April 2019

    STATUS: I'm likng the trim "Start" and "Both" options now that I have the Noise Floor setting dialed in (-54dB works for the Piano Apps I'm using). I noticed trimming the sample can make a 5 second setting produce a 1-2 second result however.

    To get long sustain's I need to manually adjust the looping of samples in AudioLayer but 5 seconds without dropping the sample works for 98% so no need to a lot of editing to make a playable instrument and turning on looping takes a few minutes for 100 samples.

    The SynthJacker "Normalize" feature helps insure I get a quality sample and I can tweak the settings in AudioLayer to insure soft/low velocity notes are good for me. I've been able to make some useful soft playing pianos with gentle "dark" tones. I'll leave "Normalize" on going forward.

    OT: I duplicate a single layer piano into 2 layers with one up and octave and one down an octave and I can see making "felt" or water damaged pianos is goin to be easy. AudioLayer lets me making some pianos that defy the label of piano.

    I tried SFZ import into Auria Pro and didn't get the file layout right... I'll try to get that right and see if this is going to be a good "buy" for Auria Pro users too. Can I open the SFZ SynthJacker produces in an editor to insure the file naming is correct?

    I think I'm ready to review this as a 5 star app. No outstanding bugs for me that I can't workaround. Too many layers of some Piano Apps cause crashes but I'm trying to make pianos that are good enough without using GB's of audio and it's working to do just that
    So, 5 layers is serving me well and I'm dropping the velocities at the very high end. If I want
    a in your face piano (lots of striking noise) I can make one in 10 minutes max. So, making instruments to my personal needs is worth a lot.

    I'd love to see updates to might include (in any order):

    wave files added as an option (currently produces AIF audio) +$2
    extend the 5 second max record to 2 digits +$3
    IAA or AB App hosting (in addition to AU App hosting and external hardware support) +$5
    run the App as an AUv3 FX product so I could send anything in AUM into it +$10

    Do all those and it's a $20 app. You might consider adding them as IAP's.

  • Now I'm working my way through my favorite acoustic instruments in BeatHawk, BS-16i, etc.
    I haven't looked at disk usage per instrument yet or how many AudioLayer instruments I can run in parallel in AUM. It's just great to have this level of productivity rather than needing to
    use a desktop/laptop solution for auto-sampling.

    That alone makes SynthJacker a must have for me and I'm getting the benefit of the $15 AudioLayer investment.

  • I'm not having any luck importing the SynthJacker sample sets with SFZ file into Auria Pro. I used a document process I found on the forum to Zip up the file
    and open in Auria but I don't hear and sound when the instruments loaded.

    Move the sfz file and the samples into AudioShare. Use audioshare to zip the folder with the files. I renamed the zip to name.sfz. Moved it to the Instruments Folder in Auria Pro. It appeared in the selection list of Lyra Instruments. Seemed to open but no sound. The same set of files works in AudioLayer (using the Name_Velocity_Note import route).

    I suspect the *aif file format might be the issue. Is there a bulk aif to wave converter for IOS? Then I could see if that's the issue. It would also be a workaround for SynthJacker not having waves file as an option. Anything to avoid using a Desktop/Laptop answer would be appreciated.

  • McDMcD
    edited April 2019

    I tried 9 of my SynthJack'ed AudioLayer instruments in an AUM session
    with Aphelian (5 MIDI Patterns counting the DRUMS) and 4 Riffers cranking out
    9 MIDI Patterns. All sounds are the 9 AudioLayers.

    I added 5 Bleass Reverbs and 3 Bleass Delays.

    DSP was about 60% on an iPad 2018 with 2GB RAM.

    That's a boat load of MIDI rendered to audio (it's most interesting from 1:15 to end when I shift the volumes to expose difference instruments:

    There's 9 of them running - all made today from AU Apps:
    Piano
    E-Bass
    Flute
    Oboe
    StringsSlow
    French Horns
    Spanish Guitar
    Cello
    Sax

  • McDMcD
    edited April 2019

    I tried another AUM project with 9 Atom Loopers playing the 9 SynthJacked AudioLayer Instruments. The Atoms all crashed around 1:35. AU's are tied together for resources.

  • edited April 2019

    @McD said:
    I tried SFZ import into Auria Pro and didn't get the file layout right... I'll try to get that right and see if this is going to be a good "buy" for Auria Pro users too. Can I open the SFZ SynthJacker produces in an editor to insure the file naming is correct?

    Yes, it's just a text file, as SFZ files are by definition. It is currently a very simple one, basically just has the sample names and their root notes, no special SFZ opcodes or anything.

    I did test with Auria Pro / Lyra early on, but now that I tried it again by just copying the files into a folder under "Sampler Instruments" I couldn't get it to stick. (They were AIFF files.) If you come up with a solution with Lyra, do let us know. Or if there is something that SynthJacker needs to do in order to make it work, I would appreciate a heads up.

    wave files added as an option (currently produces AIF audio) +$2

    Yes, this would be a good option. AIFF has traditionally been the Apple format, but maybe it has been a bit superseded by CAF (Core Audio Format), and it seems that WAVE is too popular to ignore even on iOS/Mac. My understanding is that it would also better support looping information inside the file, and would also be a better fit for possible SoundFont 2.0 support.

    Thank you for the status reports; it seems you have really put SynthJacker through its paces! Nice to hear that it seems to be working well for you after some initial hurdles.

  • @coniferprod said:
    I did test with Auria Pro / Lyra early on, but now that I tried it again by just copying the files into a folder under "Sampler Instruments" I couldn't get it to stick.

    I could ask the question again on the Auria Pro Forum. I tried some sfz's from other sources and I can't get Auria to play them. They appear to be opening but they won't play sound.
    I'm not getting sidetracked on that.

    I'll just keep making instruments for AudioLayer using AU's and hope you can add another interface method in a future update.

    I'll add my App review in the morning.

  • @coniferprod Hey, bought your app hoping to sample from one iPad to the other iPad via my iCA4+. It seems it will not support multichannel interfaces? Can you confirm or deny this? Both interfaces I have are multi-channel, the iCA4+ and the Behringer UMC404HD and neither seemed to work.

  • @gmslayton said: Hey, bought your app hoping to sample from one iPad to the other iPad via my iCA4+. It seems it will not support multichannel interfaces? Can you confirm or deny this?

    Currently SynthJacker will only use the first input of a multi-channel audio interface. This limitation may be lifted in a future update once I learn how to implement it. Since SynthJacker was originally intended for sampling external synths which are often mono, this has not been a great limitation — yet. Instrument Audio Units are sampled internally and are stereo.

  • @coniferprod thanks for the reply. I understand original intent vs. the 1000s of requests people will demand from a developer. I have not tested the AU yet. I was mainly concerned about sampling from say Beatmaker 3. I have built up a nice instrument and want to multi-sample it. Since neither of my audio interfaces seemed to work, I did find a work around. I guess in a way I have a 3rd interface, my Zoom H4nPro. Works well as 2 channel audio interface as well. So I used that and everything went perfect. If you could add multichannel inputs that would be amazing. You have already built us a gem of an app so thanks for that.

    Here is a little snippet of what I created.

  • Based on specs alone, the iConnectAUDIO4+ and the Behringer UMC404HD seem to be regular USB audio interfaces. So if you select "External Instrument", they should appear in the "Audio Input" selector. But in practice that would mean that you'd have to connect your sound source to Input 1 of the device, for now.

    I have a Zoom H1n which is an iOS compatible audio interface, and it was successfully used in SJ testing, so it makes sense that your H4nPro would work as well, but using just one channel.

    But, as mentioned earlier, I'm researching the input selection. I would like to be able to use any input in mono, or a combination of inputs like 1+2 in stereo. So I hope I can implement that soon.

    Thanks for the feedback and sharing your sounds. Happy synthjacking!

  • Happy June everyone! This is probably a good thread to announce the latest update of SynthJacker, now available on the App Store. Here's what's new in v0.6.1:

    • SynthJacker now supports the selection of audio input(s) for external audio interfaces with more than one channel. If the audio interface has an even number of channels, they are grouped into stereo pairs. Single channels may be selected individually. Recordings from an external source are stereo or mono depending on your input selection.
    • Internal recordings from instrument Audio Units are always stereo 44.1 kHz.
    • The SFZ file generated by SynthJacker now contains velocity information for the sample regions.
    • SynthJacker now generates a standard MIDI file of the selected sequence as a side product of the recording.
    • Some instrument Audio Units showed a blank view. This has been fixed.
    • The virtual keyboard no longer obscures the instrument view.
    • You can create a copy of any sequence by swiping to the left in the master list and tapping Duplicate.

    Go forth and multisample!

  • @coniferprod said:
    Happy June everyone! This is probably a good thread to announce the latest update of SynthJacker, now available on the App Store. Here's what's new in v0.6.1:

    • SynthJacker now supports the selection of audio input(s) for external audio interfaces with more than one channel. If the audio interface has an even number of channels, they are grouped into stereo pairs. Single channels may be selected individually. Recordings from an external source are stereo or mono depending on your input selection.
    • Internal recordings from instrument Audio Units are always stereo 44.1 kHz.
    • The SFZ file generated by SynthJacker now contains velocity information for the sample regions.
    • SynthJacker now generates a standard MIDI file of the selected sequence as a side product of the recording.
    • Some instrument Audio Units showed a blank view. This has been fixed.
    • The virtual keyboard no longer obscures the instrument view.
    • You can create a copy of any sequence by swiping to the left in the master list and tapping Duplicate.

    Go forth and multisample!

    Awesome - thanks!

  • I hope the SFZ outputs work in all the DAW's that accept SFZ input. If you know of one that does comment here and we can test them.

    For AudioLayer it creates filenames that import well.

    I've great that it can import into NS2 and tried but didn't figure it out completely and stopped trying.
    Maybe the AudioDabbler can add a few more steps into his video. I haven't mastered the NS2 UI but there's @ScottVanZandt amazing proof that it's 3 layers of sampled velocities can make great music:

    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.

    If 3 layer samples can do that then I want to figure that out (at some point). I still muck about in AUM so AudioLayer is needed.

    It's also a clue to Obsidian's unique capabilities as a synth/sampler hybrid. The audio examples of the IAP's don't prove the case for me but this track blows me away. Scott posted that it looks weeks to make all the samples but now he has the IOS orchestra at his disposal and the great NS2 workflow.

    I think StageLight imports SFZ's and maybe Audio Evolution Mobile Studio too.

    In this new version the max sample time has been upped to 10 seconds and I didn't know about the MIDI file. There's also an option to save the whole recording of all samples as a wave file which might have some applications for importing into apps like SpaceCraft granular or anything that can chop up audio detecting transients (Samplr, ReSlice).

    The looped back audio works well for sampling IAA Apps like Animoog if you have an audio interface. If the max sample time could be increased to minutes there might be some use cases for "sampling" YouTube videos or other IOS Apps that only produce audio out. The digital-to-audio and audio-to-digital
    artifacts will be a function of the quality of the Audio Interface's implementation. Mine is a $200 Presonus and I'm very happy with the lack of noise or any changes to the audio.

  • Some additional information: I didn't single this out in the App Store notes, but I've tested both Factory and Aparillo from Sugar Bytes, which showed a blank screen in SJ v0.5.x, and now they work fine.

    Over at the Blip Interactive forum people seem to have been pleased with the Obsidian compatible filename options. Regarding SFZ, the output from SJ 0.6.x has been tested with Auria Pro / Lyra, but if there's any weirdness, let me know (preferably send feedback e-mail from inside SJ).

    I can most empathically tell you that the individual note duration will never be increased to minutes. You'll just have to use something else for recording YouTube videos, since the point of the app is to record multisamples of synths.

    For sampling a piano, it definitely works best if you do it in multiple runs, because the complete notes have very different lengths (low notes are longer, high notes are brief). Also, without automatic loop detection piano samples will never sound as good as they could. But maybe, just maybe, that could be addressed some time in the future.

    Also, I don't know which will happen first: being able to use KORG Gadget instruments, Thor and Animoog as AUv3, or supporting Inter-app Audio SynthJacker. Or neither. :smile:

  • OK. Minutes is too big of an ask... how about 20 seconds as a max in the next release so I can avoid hitting the looping in audio layer for a few instruments.

    NOTE: It helps of a few people say they would buy the app if it recorded samples of some requested maximum length.

  • edited June 2019

    Please excuse my ignorance, but I'd like to know how I might use Synthjacker.

    I don't have any external synths to "jack", but I see you can use AUv3 apps to create the samples to be imported into a sampler.

    The only standalone sampler I have is Chameleon. I don't have AudioLayer. I've got the sampler in NS2, Auria Pro, and I think a couple others, and Samplr of course.

    However, I don't understand a user scenario where I'd use Synthjacker. I won a redeem code for it is why I happen to have it.

    So, I select an AUv3 synth. Lets say DRC, pick a preset and them sample it with Synthjacker. I can then import those samples into any of the samplers I have, or into Chameleon.

    But, I can sample AUv3 synths into any of those samplers directly without Synthjacker. I can play a note into Chameleon and create stacked presets from several inputs and AUv3 synths.

    In what scenario could I benefit from the use of Synthjacker instead? All of the scenarios for Synthjacker I can think of seem redundant from what I can already do with the apps I already had. I'm sure I must be missing something though.

    Any suggestions of unique and creative situations where I'd want to use Synthjacker as opposed to what I already have?

  • @skiphunt said:
    Please excuse my ignorance, but I'd like to know how I might use Synthjacker.

    Synthjacker automates the creation of sample sets that have a lot of files. You can create a sample set by selecting:

    Up to to 127 different MIDI controlled notes. I typically sample from C1 to C6 with 4 notes per octave so
    that's 21 notes to be sampled

    Then I like to sample a variety of Velocities like 20,40,60,80,100 so that multiplies the number of samples to 105 output files.

    So, I load up an AUv3 instrument (iSymphonic, Colossus Piano, Ravenscroft 275) and import the resulting folder right into AudioLayer and have an AUv3 clone of a complex AU app.

    If you make sample set with just a few samples then it's not saving up a ton of time.

    Chamelon, by the way, supports a maximum of 15 samples. For me it's a toy. But for some that might not be an issues if they want a decent drum kit or maybe some longer samples triggered by MIDI Notes.

    SynthJacker also generates SFZ files to allow importing into Auria fast and get some great new instruments that way.

    I've read that it helps with NS2 workflows too... I just didn't get it figured out and gave up trying. I'll re-visit that someday because NS2 will add audio tracks and I might move in and learn it.

  • edited June 2019

    @McD said:

    @skiphunt said:
    Please excuse my ignorance, but I'd like to know how I might use Synthjacker.

    Synthjacker automates the creation of sample sets that have a lot of files. You can create a sample set by selecting:

    Up to to 127 different MIDI controlled notes. I typically sample from C1 to C6 with 4 notes per octave so
    that's 21 notes to be sampled

    Then I like to sample a variety of Velocities like 20,40,60,80,100 so that multiplies the number of samples to 105 output files.

    So, I load up an AUv3 instrument (iSymphonic, Colossus Piano, Ravenscroft 275) and import the resulting folder right into AudioLayer and have an AUv3 clone of a complex AU app.

    If you make sample set with just a few samples then it's not saving up a ton of time.

    Chamelon, by the way, supports a maximum of 15 samples. For me it's a toy. But for some that might not be an issues if they want a decent drum kit or maybe some longer samples triggered by MIDI Notes.

    SynthJacker also generates SFZ files to allow importing into Auria fast and get some great new instruments that way.

    I've read that it helps with NS2 workflows too... I just didn't get it figured out and gave up trying. I'll re-visit that someday because NS2 will add audio tracks and I might move in and learn it.

    Let me make the scenario simpler.

    Let's say I want to use the AUv3 synth DRC in Auria Pro. I can use Synthjacker to create up to 127 different midi controlled notes (as you say). Then, I can quickly import SFZ files into Auria (as you also say)

    Now, I've got these samples imported into Auria's sampler and ready to play DRC in Auria.

    Why would I do this when I can already simply use the DRC AUv3 synth in Auria without making any samples at all?

    I get being able to combine several synths into one complex AUv3 synth. But, I can more or less do that already with just Chameleon.

    I'm not doubting there's great use to be had with Synthjacker. Otherwise, the developer wouldn't have created it. It's just that I'm having trouble seeing how I can use it to benefit at all... considering I can just load up 3 or 4 (or more) synths in apeMatrix or AUM/AB3, etc. and have the result of a complex synth sound too. Where does the benefit of creating all of these extra sample files to import into another sampler factor in?

  • @skiphunt said:

    There's lots scenerios where it'd be useful, but i'd mainly use it for sampling hardware first and sampling desktop synths to the ipad. Sampling an AUv3 you already own could reduce cpu in projects, or you could be sharing a project to someone who doesn't have the synth etc.

    If you just want to use an Auv3 as it is you already own, there's not a lot of point in sampling it, but maybe you could create a hybrid instrument with different synth sounds in the different layers for example :)

Sign In or Register to comment.