RatFace

I can't directly help with the original question, and neither have I done more than skim the previous messages, but: I was playing and recording (so MIDI over USB to my phone running Audio Evolution to do the midi->audio conversion (which is generally extremely good). Right at the end of playing, I got a stuck note, and thought "duh, I'll have to do that again"... BUT.... when I stopped recording (though before stopping I reversed the bellows direction, which sends all-off which cancelled the note) and played it back, the note wasn't stuck! So in this case, I don't think the stuck note was caused by my code running on the Teensy - just be aware that the midi->audio conversion system might be buggy instead.

FWIW, I got one of these to use with my newer phone (which doesn't have a headphone socket). The audio quality out is absolutely awful, and I couldn't get it to talk to the Teensy over USB at the same time as the audio (or maybe it was charging and something else?). I reckoned that trying to find one of these things that actually works (though just power + USB is probably OK) is probably more expensive than finding an old Android phone!

Actually, I do have a separate battery pack (4x AAA) for the audio amp. The phone just has to power the Teensy and itself. In practice, my guess is that the main drain is the audio app and the display - though actually I can turn the display off and the audio app continues to run... until it doesn't (which can take quite a while).

Yes. And you can have separate sound fonts for each track too. The problem is that the phone's usb port is already connected to the Arduino/Teensy to (a) power it and (b) do the MIDI thing. I might be able to power the Teensy separately. Maybe it would be possible to wire things up to allow the phone to be charged separately from its usb connection to the teensy...

I may have mentioned elsewhere, but I use Audio Evolution Pro (AEP - https://www.extreamsd.com/ )running on my old phone (Pixel 3a) for my midi bando-thing: I have a project saved that is "ready to go", so I don't need to do any setup when I want to play Having said that, I do need to poke it, so it's not a fully embedded solution: Turn the phone on Start the app Load the project Arm the two tracks The phone is attached to the box - no wires to some external units etc. No need for an external power supply. The audio output (requires headphone socket!) goes back into the bando-thing, into a little 5V amp and then speakers This lets me use a custom soundfont (sampled bandoneon, plus also I put piano and a few other things in there) I also purchased the Toneboosters reverb which is pretty good I can use the metronome facility in AEP I can make use of effects, mixers, panning, multi-tracking etc At any time, I can just hit record and record my playing, then listen/review etc AEP has its own audio drivers - the latency of these is extremely low (unnoticeable) The audio quality of this setup is (IMO) amazing - it's infinitely better than what I was getting initially with fluidsynth. The cost (second hand phone + software) is about the same as the Pianobox mini ii. The functionality is greater than what that box will do. The only downsides are: My setup can't power the phone (actually, the phone powers my bando-thing!), so I'm limited to about three hours of play time. I wouldn't want to be on stage and changing settings on a phone app! You can't embed this into an instrument and have it automatically boot up and run the app with the right project (I've wondered about suggesting this to the developer!) Example and photos: https://photos.app.goo.gl/szQRu4fn9tfmPcSL9 In fact - this wasn't meant to be a "proper" recording - I'd been practicing a few days ago, and it's just so easy to "hit record" that I did, and afterwards thought "that's a keeper" Edit: I just noticed that all the photos in that video are before I settled on this setup, which shows the phone mounted on the box:

By the way - when I printed mine, I used "strong resin" because I was worried about the thin stems being strong enough. However, I wouldn't do that again. Being so strong, they don't flex, so there isn't much room for error! Luckily I got the dimensions right, though a couple of them cracked when fitting. I think a more regular (and cheaper!) material would have been fine. The main thing, though, is that I think the resin dust was somewhat irritating to my skin. I don't know for sure that it was this - it might have been the wood finish that was taking a long time to cure. Either way, it took quite a few months for me to not feel there was some irritation going on.

The ones I used were these: https://www.amazon.co.uk/dp/B07X3T14CV/ref=pe_27063361_487055811_TE_3p_dp_1?th=1 You can reduce the size by 1mm or so on each side by trimming off the plastic. One side of the switch is actually just empty plastic. It might be possible to trim this off, saving around 3mm or so, but you'd have to see whether that would require the switch case to be glued. I used them for a bandoneon, and even there the spacing was tight. You could (perhaps) mount the switches in two layers, allowing them to be staggered. Regarding printing - I got mine printed through https://craftcloud3d.com/ The difficulty there is that you have to be confident you got it right first time!

Did you check the switches still activate ok? If they're disappearing enough to stop you getting the key cap on, doesn't that mean the key cap will limit the travel?

Thanks. Yes, they sound nice together. I did cheat slightly in the recording/mixing by boosting the guitar mic slightly - the live sound had an acceptable balance (I believe), but I would have preferred the concertina to be a little quieter. Me too! I am part way through transcribing the parts in the vain hope that if they're easier to read they'll be easier to play. The edition that I have (emailed to me) has got pencil markings (fingerings etc) all the way through suggesting the previous owner had made a serious attempt to learn/play it. On the cover there are two stamps: Presumably the crossing-out indicates that it passed from Mellor to Maddocks. Harold Mellor gets mentioned here: https://concertina.org/wordpress/wp-content/uploads/2021/01/PICA-Volume-4-2007.pdf and it appears that c.net member @Gerard owns his concertina: and HC Maddocks here: https://concertinamuseum.com/NC50201-003.htm I don't know which one of these two was responsible for the pencil marks...

This is a really pretty piece, and sounds lovely. A splendid concertina too! I did nearly give me kittens though: I had just uploaded my "Andantino" when I saw the title page of yours pop up as I finished test-watching mine and I thought you'd gone and beaten me to it!

I had been corresponding with Scottish guitarist Stewart Kelly about music by Regondi, and then we realised he was due to give a concert down here in Oxford, so we decided to play a piece by Regondi together:

Yes - exactly! There are USB splitters (into headphone + usb), but I suspect the usb side is then for charging only: Maybe another option would be to use a small USB hub (one in from the phone, two out)... I have no idea if it will work. Though perhaps I can try with my current phone. Apologies for the thread drift

I really like having the instrument completely stand-alone - so no need for external power or devices, and wired headphones are not at all inconvenient. I clip my phone on the top - it takes in the midi, gives me far more control over the audio generation than I really need(!), and also powers the microprocessor over USB (with the phone display off, it will last a few hours). Then I have a little battery powered amp/speaker system. I took it on a train journey recently, and had an hour or so to kill at Basingstoke, so was able to play (silently) on the station platform. I didn't even get arrested! I do have a problem/question about that though - a lot of newer phones don't have a headphone socket, so I'm unsure how to do this on a phone that only has USB. I need the USB to power the Arduino/Teensy, and receive midi, and output to wired headphones. Does anybody know if that's possible (or suggest an alternative - I quite fancy getting a new(ish) Pixel, but this is stopping me!)?

Yes - typically the chain is something like: The keys will be wired to A programmable microprocessor - Arduino. This reads the keys (but often has to do it by scanning the keyboard, wired as a "matrix", which can take some time). It will also read things like a pressure/force sensor, which will have a clock (typically 80Hz). On mine, the processor completes its loop 80 times a second - so that's 12ms latency straight away! The microprocessor sends out midi: That might go direct into a hardware midi expander box. I expect this results in the lowest latency, but I don't know/think any/many allow for custom soundfonts Alternatively, it might go to an external computer/phone over USB (fraction of a ms) which is converting the midi into an audio signal. Audio Evolution Mobile Studio on my phone - it has custom drivers to reduce latency, and supports custom soundfonts and effects etc. I don't know what people use if they plug into a PC. I think some people do this using something like a raspberry pi inside the instrument - which is basically the same thing. Alternatively, you could synthesize the audio on the microprocessor (avoiding midi altogether)- either by writing code to load/play soundfonts, or by just building the audio out of a truncated fourier series. I quite fancy doing the latter - the sound wouldn't be "real", but it would be very convenient (no external computer/phone needed), and you could still send out midi for recording etc. Then the audio signal will go straight to wired headphones/amp+speakers. Bluetooth has a minimum of 40ms, which is a lot. There may be other ways of doing things.

I think a real concertina starts speaking immediately the buttons starts going down - so I'd be inclined to trigger the note as soon as possible to emulate that (especially as key switches only trigger after a few mm travel anyway). The other thing is that there will inevitably be delay in the system anyway - through processing on the microprocessor, and then in conversion from midi to audio. I suspect that delay is more than you want anyway, and you wouldn't want to add any more. I play mine through my phone (Pixel 3a) into speakers/headphones, and the processing delay isn't noticeable. However, if I use my tablet (Galaxy Tab S7 FE), the latency is awful (though I didn't notice it at the beginning when I could really play yet). I think if you made an instrument that speaks too quickly (i.e. faster than a real concertina), then that's massively better than going the other way and running the risk of adding too much latency...