The Sound Tech Hearing Aid

 An acquaintance of mine recently got a hearing aid and complained about the loudness. Let's face it: many of us, if we did not take care of our ears well enough, may need hearing aids at some point in our lives. So, why not seize the opportunity to make a fun and useful gadget?

In my opinion, the ideal sound tech hearing aid would come with the following features:

1. A decent mic, maybe e. g. a DPA omni capsule or similar.

2. A pad button (e. g. for rock concerts; optional)

3. A gain button

4. A noise gate (to sort of mimic the cocktail party effect; or an EQ preset for better voice recognition, maybe enhanced by AI if the noise is of the same frequency spectrum as the voice(s), plus giving the voice(s) a +6 dB-ish boost. The AI feature could be an optional add-on, like in a lego toolkit)

5. A compressor (e. g. if you are listening to a conversation where a speaker always goes quiet at the end of their phrases) with automatic makeup gain

6. A mute button for when you want to be left in peace

7. An EQ with at least 6 frequencies - for example, give the higher frequencies a mild boost if you lose them as you get older. Also, some people have individual "valleys" in certain regions. You could establish an individual preset in collaboration with a doctor and/or hearing aid tech based on the results of your hearing test, with different settings for each ear. Usually, you would want to go for a clean dB(A) curve (or maybe less bass for when you are not working / listening to music) - taking the Fletcher Munson curves into account (iirc, they mostly look similar at different sound levels, but maybe the best response would require a real-time analyser, and somewhat shifting EQ curves based on the current avg dB level? On second thoughts, this would likely feel weird, as I would prefer one fixed setting - probably best to settle on a compromise).

8. various FX (optional, for fun)

9. A signal splitter for sampling interesting noise where the 2nd pathway goes to an external data storage device (for example: smartphone memory card) and saves the signal in a lossless format (pre or post eq, fx; I would prefer .wav). The data can then easily be extracted into your OS / DAW of choice (Maybe, to prevent abuse of this feature or spying, the hearing aid should feature a red blinking LED, like an earring, whenever it is recording)

It could e. g. be controlled by a smartphone app that is linked to the device via bluetooth and a password.

*It has since been brought to my attention that devices like this exist, but most of the features seem to be automated. I would like to be able to operate them manually, as I may need a different preset when I am at a party where I would like to listen to people vs. when I am at a classical concert and would like to experience a wide dynamic range (no filters, everything flat). Also, sadly, hearing loss can worsen, so adaptations to the settings may have to be made. A sound tech could do this independently between doctor / hearing aid tech appointments.

One more idea:

HRTF simulation with two e. g. hypercardioid mics

Imagine that you have two mics, one per ear. Let's assume that both are facing forward like your eyes, and thus have less reception from behind, similar tothe way human ears work. 

Due to time and loudness (ITD, ILD) differences between the ears, plus the fact that sounds from the front will always sound sharper than from behind, it should be possible for an algorithm to place the sound correctly in a 360 degrees circle (let's start with the 2D, 3D may be more complicated or require a third mic) with the human brain as the center. Once this position is established, there could be a postprocessing filtering curve that corresponds to the average human HRTF for both left and right ear (the curves may differ from ear to ear), thus simulating binaural hearing, making the sound more natural and allowing the user to precisely locate any sound, just as before with the "natural (g)ear". There could be a lowcut for this feature, i. e., it need not be applied below 500 Hz. As the processing requires calculating power, the bar could be set higher, maybe starting from e. g. 1 kHz - this would have to be tested.

The ceramophone

 I recently discovered by accident that my favourite coffee cup can produce two different frequencies, and a chord containing both.

I test-played it (explanations in German, maybe automatic subtitles work) here:

Reel 2
Reel 1 (and comments)

The most important observations from these experiments are:

- My cup produces the frequencies F6 and Eb6.
- The handle adds mass on one side, which causes an asymmetry, and introduces two different frequencies. Near the handle, the extra mass limits the vibration, causing a higher note, F6. On the opposite side, the cup can vibrate more freely, creating the Eb6. According to ChatGPT, this means that the cup has two orthogonal modes. Without the handle, both frequencies would be the same.
- In the middle (in my case, at ~30 degrees away from the handle), you can PLAY A CHORD. The intensity of both notes varies as you move along the cup!
- Angle, vertical position and force of the nail / snipping influence the sound. For reproducible sounds, playing as horizontally as possible worked well for me.
- Open sound vs. dampening sound: To create a clear, bell-like sound, play the cup on a surface. To make the sound very short, hold the cup in your hand. To dampen the sound early when playing on a surface, simply touch the handle or rim of the cup, as in any other instrument.
- Further modification: Fill the cup with water to make it sound lighter. For lower tones, take a bigger pot such as a plant pot with a handle on one side, or add some weight on one side, try a parabolic shape with a flat bottom, as in my cup.

Why is this interesting?

The cup offers two frequencies that are not observed in natural harmonics of a guitar string, where only fifths and thirds are observed. By using ceramics, you get more and varied sound colours.

What can you do with it?

In theory, by experimenting with cup size, cup shape (different conic sections), cup wall thickness and handle mass, you can create any interval from a minor second to a major seventh. I think that the wall should not be too thick. My cup has approx. 3 mm wall thickness. (Gmundner Keramik Kaffeetasse Maxima, rainbow edition.)

You could then introduce something that makes the snipping process (and the resulting frequency / frequencies) reproducible, like a hammer as seen in a classical piano.

If you build the hammer so that it can rotate 180 degrees to and fro, you can start with one note, gradually move into the chord, first hearing the first note more loudly, reaching equilibrium, then hearing the second note more loudly, until you end in the second note.

Imagine that, for example, two cups are played simultaneously. In jazz music, two notes of any chord produce the lead tones of this chord. (2-7) Let us imagine that you are playing the lead tones of a chord of your preference. Now, you want to change to another chord. You could do this by making a simultaneous rotation of both hammers. For a while, you will have very colourful chords, until you end in the final tones, which could be another lead tones chord pair. It would be similar to a guitar slide, but in the form of a chord, and in stakkato mode. Maybe you could pre-program the board to make the slide in a specific amount of time, and hold a predefined rhythm (sort of like a step sequencer), and map this transgression on a custom button. You could use this to accompany another instrument. Since the notes are not very long, you would probably want to play them more than once, maybe as quarter notes. In open playing mode (on a surface), the maximum length of any given note is perhaps approx. half a second.

You could also play e. g. two cups simultaneously in chord mode and get e. g. a Dm7 chord with all four notes. Some mathematicians, musical theorists etc would need to think about which chords to choose, how to position the cups and operating knobs so that it intuitively makes sense to a piano player / organist / xylophone player. With 6 cups, you could, with good planning, represent all the notes on a well-tempered keyboard. You may want more than those, though, to have more octaves, and perhaps to try more colours / different intervals.

How to operate the instrument:

Such an instrument could be operated with knob potentiometers which you recognise from every standard mixing board. The two notes would be written on each knob so that

0 degrees = -infinity = frequency 1
12 o'clock = 90 degrees = chord
180 degrees = max = frequency 2

Turning the knob will make the hammer move to the indicated position.
To play the note(s), press the knob from above = the hammer hits the cup.

This would allow you to play frequency 1, move to frequency 2 and move back to frequency 1 again without having to play any position in between on the way back to frequency 1.

Further useful features:
To link any knobs of choice, so that chords will be played (up to 5, for e. g. Eb9).

A mute button for knobs that are linked, so that they can be left out occasionally, or during transgressions when the sound would get too messy.

Two kinds of dampening the sound (could be solved by introducing foot pedals):
1. making it lower and shorter by lifting the cups in the air
2. shortening the note by touching the handle or rim

What does it sound and look like?

My cup sounds similar to a bell, very clean. I imagine an instrument of this sort to be rather big, at least the size of a marimba.

As far as I could hear until now, external conditions have little influence on the pitches of the frequencies, as long as you play it at room temperature and not e. g. in an ice cave.

The disadvantage is that every cup is probably limited to its own specific frequencies, which means that any instrument would have to be pretty big. Still, I think it could be fun to play.

How to mic it:

I would imagine the operating surface with the knobs similar to a mixing board table and the cup mechanics in a box underneath. Ideally, for use on noisy stages, the cups should be acoustically isolated from the outside. A mic, probably a small membrane mic, or two as for a piano, could then be placed inside the box. For practical reasons, any touring musician would probably have two fixed small membrane mics with XLR outs pre-installed in the box.

The bidirectional step sequencer

 I built a bidirectional step sequencer in PureData, based on a 4-step sequencer tutorial I found at GitHub. Basically, you can rebuild it and use it by following the instructions, changing the parameters as you like.

Here is how it works, more or less:

The heart of the machine are the vertical sliders, each representing a MIDI pitch value which is sent to a number box and converted to a corresponding frequency. Whenever a bang is triggered (the number box above the select 0 1 2 3 box gives the select box a number ranging from 0 to 3, and the select box specifies to which position the bang is directed - slider 0, 1, 2 or 3), the float value of the corresponding slider is fed to the number box below, from where it is converted from MIDI to frequency, and being output as a sine wave.

Now, the machine has two step counters, a forward and a backward step counter. While the backward counter counts down from 3 to 0, the forward counter counts up from 0 to 3. At the end of the cycle, the counter is reset to the initial value (the inlet of the respective float). Also, a number is added to the cycle counter of the side which is currently running. After a given number of cycles (here: 4 cycles), the machine switches to the other step counter. The given number of cycles triggers a bang. This bang in turn triggers two bangs. These two bangs trigger two switches, of which one is always turned off and the other always turned on, which means that the running direction of the sequencer is switched. Whenever the bang that switches direction is triggered, the cycle counters are reset to 0.

You can change all of the parameters: speed (specify the ms as the number after metro), number of notes (add sliders and change values (do not forget to change all related arguments!) as needed), frequency range (adjust the settings of the vertical sliders if required), number of forward and backward cycles before the jump (change the sel argument of the cycle counters, try asymmetric setups if you like, for example make one metro slower/faster than the other, have more cycles in one direction than the other).

You could also change the sel arguments, e. g. if you want the sequencer to run only 3 notes when running forwards, but 4 notes when running backwards - then you would change the step counter sel argument to 2. It will now reset the step counter when reaching the argument 2 (after 3 notes instead of 4).

If you want the sequencer to run only in one direction, choose the counter you like better and remove the following: switch bang, tbb box, bang to reset cycle counter, the whole cycle counter plus the whole dismissed side of the machinery. You are left with one counter (either backward or forward) that resets to the initial value at the end of each run, therefore running as long as you like, at the metro tempo specified.

This is an open source machinery. If you use it, do buy me a coffee or beer if we ever bump into each other. Thank you.

Bidirectional 4-step sequencer with automatic direction switch after 4 cycles. You can also switch at any given time by clicking on the direction switch (uppermost bang) manually.

You can also build a setup that allows you to switch directions manually at any given time, see below:

Bidirectional step sequencer with manual direction switch (click at any time)

Additional versions (see pd directory in the link below and open the files that interest you directly in pd. The setups are getting too big to make screenshots...):

Bug fix: Introduced a short break before direction switches, so that the same note can be heard twice as two separate events.

Made a version with two limping metros. At present, the metro is symmetrical, so that the same note lengths are played in either direction. The metros are independent of each other and can be changed accordingly.

The palindromic beat:

Another idea is to make a palindromic beat, e. g. the classical

Prim - Third - Fifth - Third - Prim, 

for example, in the form of C-E-G-E-C.

In a sequencer, this can be reduced to four notes, with e. g. C on the first beat: C-E-G-E

Let's assume that we would like to play the figure

1      2      3     4      1

C     E     G     E     C

Now, a fun thing to do could be to cut the palindromic sequence somewhere in the middle and starting e. g. with the top note, that is

GECE

and let it run for a couple of loops. In this case, GE (3 and 4) would be the intro. After a few loops, the sequencer switches direction to

ECEG

and again, after the first E, you are back to the beat no. 1, and the melody is looping as before.

However, you may want to make each transition (transition 1 and 2: from GECE to ECEG, and from ECEG back to GECE) in such a fashion that it can be done in the same amount of notes as notes in the step sequencer. For simplicity, let's say that we want the change to be done in 4 beats.

The easiest way that I have found to do this is introducing filler notes. As you can see, the first sequence GECE ends on a 2, while the mirror sequence ECEG starts on a 4.

So, by introducing one note after the end of the first sequence, for example D#, you can continue on 4 and are back in the original palindromic beat after that. (Note: Do not forget to adjust Limping Metro delay times to fit the original rhythm pattern, if you wish, i. e.: the delay times have to be mirrored as well.)

Similarly, as ECEG ends on a 3 and GECE starts on a 3, by introducing three filler notes, e. g. A A# A on 4, 1, 2, you can continue on 3 again.

A bidirectional 4-step sequencer with 4 cycles, a palindromic beat, a limping metronome, and filler notes

The setup could be nice for an impro where just like a human percussionist, the sequencer may want to do a little impro before going back to the original rhythm. The metros are also independent of each other, so one side could be sped up compared to the other. It may also be possible to transpose one side (to symbolise key change, as could be useful when switching between verses and a chorus), but the more practical thing might be to use one cycle counter to trigger a sequencer with new notes. I see the most potential of this application for longer verses, intros or jam sessions.

Here are my pd files to download and modify as you like.

Make Dante Controller - Device safety pincodes global

 Imagine the following situation:

You run sound infrastructure at a theatre, broadcast company or other venue which often hosts guests. Your Dante infrastructure consists of two types of devices:

a) fixed devices with patched connections that should not be touched by anyone (amps, loudspeakers, a Vivace system, wireless mics, a stagebox, a Sennheiser MobileConnect station, etc)

b) devices which other people may want to, or need to, connect to (mixing board, maybe a Mac laptop with a DVS, maybe a video beamer, etc) in order to either stream their content (backing tracks, videos etc) or make a multitrack live recording

How do you make the system as safe as possible while allowing for enough freedom to use it as you and other people see fit, without any limits to creativity?

The expensive solution is the Dante User Manager, which is not affordable for smaller theatre venues.

If you have a Dante Controller, there is an option to lock devices which should not be messed with by typing in a safety pincode. If you want to change any patches on those devices, you first need to type in the pincode, and only if this pincode is correct do you get access to the Dante device.

So far, so good. However, if Dr. Evil tags along, has a Dante Controller installed on his laptop and patches his network cable into any of your Dante switches, he can, again, access ANY of your Dante devices on his own Dante Controller, from his laptop - with the potential to wreak havoc on your entire infrastructure.

There would be an easy workaround:

Make the safety pincodes of locked Dante devices global. If any Dante device is secured with a pincode, no one can access it again without entering said safety pincode. (Perhaps there should be a smart reset function as well, in case people forget passwords, like resetting the pincode by sending an automatically generated reset pincode to the e-mail account registered with the Dante Controller.)

That way, people could patch their connections e. g. to the mixing board, but you would feel safe knowing that they cannot access e. g. the main amps and change the output.

Bonus feature (maybe for a small fee): Enable locking of individual channels, globally. This way, a guest may be able to patch their backing tracks e. g. on a Stereo LR track channel 31 and 32, but not on, say, channel 1 and 2 which maybe goes directly to the Main LR PA.

I believe in open source technology wherever it is accessible, and while I am grateful that Audinate provides free training for their certificates, it would be great to provide a downscaled version of the User Manager for free or a smaller fee which is affordable for smaller venues (like an annual subscription for ~50 USD - maybe, since theatres are public venues, with a sponsoring partnership, i. e. putting the Dante logo in some official theatre publications, or similar).

Do share this with other people if you think the idea is useful.

The MIDI driving wheel / car pedals drumset

 

A car drumset... ?

A friend of mine, Martin P, shared this photo on facebook (source unknown). This led me to the following product idea:

The MIDI driving wheel drum set

Could be produced by: Doepfer et al.

How it would work:

The MIDI controller itself is a cloth with MIDI buttons that can be placed over a driving wheel / over the pedals. Regarding the pedals, light tapping would have to suffice to trigger a sample. No pedal function on the brake, for safety reasons.

This controller would be linked to a set of drum samples which are individually mappable and can be customised / replaced at any time (e. g. for left-handed drummers, custom drum setups people are used to, special cymbals or other special percussion elements that aren't standard).

The audio signal would then be fed to the car's stereo. The MIDI signal could be fed to a MIDI-capable interface to record any performance to a DAW.

There should be a global mute activated whenever any gear and the motor are active, so that playing would only be possible if the engine is turned off or idling. Additionally, incoming phonecalls would need to have priority over the drumset.

As some smart people have pointed out, wheel rotation is an issue. The most convenient workaround could be a bluetooth device to transfer MIDI data to a bluetooth receiver - and from there, MIDI could go to an interface / DAW, and the sound would go to the car speakers, similar to mobile phone calls. Latency could be an issue. However, I am confident that this will get better with newer encoding standards and technology in general improving all the time.

I see a big market here - for people who love drumming but don't bother to build this by themselves.

If you use the idea, please give me credit and show me the product when it is in its final stage.

Thanks.

Universal wireless audio signal codec / data transmission protocol

 For the most part, wireless audio signal transmission applications are provided by two competing companies starting with the letter S.

Each has their benefits and drawbacks. But now, let's have a naive dream.

Make them talk to each other.

Let's have a universal wireless audio signal codec / data transmission protocol, similar to the MIDI standard of synth communication.

There are many situations when this would come in handy. As the market shares seem to be about 50/50, people who use S1 will often bump into S2 users, and the logical thing would thus be to enable both systems to talk to each other.

Examples:

A theatre uses a S1 in-house system. Guest artist who visits brings their S2 headset and system, but wants to have a spare headset provided by S1 as a back-up. To facilitate a quick headset change, it would be smart if the S1 was running the entire time and the S2 could dial in to the S1 system. Granted, thorough frequency management would probably be sufficient, but in times of narrowing frequency bands, more efficient use of available frequencies, by using the same codec / transmission standards if needed, could help make things run smoothly.

A low budget theatre inherits four transmitters from S1 and four from S2, plus a receiver unit from S1. Maybe the S2 unit happened to break down or was sold off. Now, they have a production with seven leading roles who require headsets. It would be beneficial to be able to use all the transmitters with the S1 receiver.

Another theatre has been using S1 gear for a long time, but due to some special features which are only provided by S2, they want to switch to S2. For sustainability reasons, however, they still want to use their older S1 gear in combination with S2. Instead of having to throw everything away, they could use the universal codec mode.

Obviously, this can only be applied to future devices. Those should have

- their individual firmware (either S1 or S2) to ensure downwards compatibility
- the new universal transmission standard
- the ability to send both signals in an encrypted mode
- cross-compatible re-chargeable batteries. It has happened that guests with an S2 system had short-lived batteries with no replacement at hand. If those guests could be helped with long-lived re-chargeable batteries from an S1 system, that would be a big help.

How could this be achieved?

- The new compatibility mode would require both companies (and any other competitors) to send standardised data packages which transmit signal / battery information etc.
- Encryption would have to be standardised, and used.
- An Universal ID allocating / syncing system would be required.
- In compatibility mode, any device needs to recognise and allow "foreign" devices to send or receive signals.
- Ideally, I would envision a universal protocol that can run on different codecs and automatically recognises which codec is used, so that multiple devices can talk to each other. For example, if you are using a state-of-the-art transmitting unit, the Universal Protocol should run on, and recognise, 16-QAM. However, if any of the devices that need to talk to each other has an older / less complex codec, like only using phase shifts, the Universal Protocol should automatically switch to this older standard for the oldschool transmitter, and use an algorithm to avoid any data transmission conflicts.
- Latency should be 3 ms max., as usual.

Obviously, the Universal Mode would limit the range of special functions that could be provided by system-specific applications. The sound may be different, as different encoding systems apparently make a difference. However, I can see thousands of potential situations where compatibility between devices from different companies would be incredibly beneficial. In the age of the climate crisis, any approach to increase longevity of existing systems and use fewer resources / produce less new material is a big plus. Green Technology will be increasingly embraced by Early Adopters and politics, e. g., in the EU, by governmental bodies, and so on - it could be possible to raise substantial funding for this idea.

I do not think that this idea would threaten either S company's sales. Both systems have their advantages, both work reliably and safely, and happy customers will usually stick with one of the two. It will remain best practice not to combine the two (or more) if it can be avoided. However, it would be a shame to miss out on the advantages an Universal Mode could offer. Most importantly, increased flexibility of frequency use at a time when available frequency bands are under constant threat of further narrowing - especially considering the fact that wireless audio signal transmission purposes are classified as Secondary Users in Austria, with Primary Users and their interests having priority over us.

Looking at mobile phone chargers, the EU has issued legislation to standardise chargers and charging slots, which has made enduser life a lot more convenient, yet hardly affected mobile phone sales. A similar form of legislation could require the Universal Protocol to be put into practice. Ideally, I would imagine it to have global compatibility. Similar to the MIDI Association, the basic foundations could be laid out at the next WRC.

To conclude, I hope that some smart brains consider this idea and will put it into practice. My main motivation behind this suggestion is to see more of a Green Event Industry taking forms, to make the AV industry more sustainable wherever possible.

Diploma thesis

In June 2020, my study mate Maria and I recorded our diploma thesis at the tone-art studio.
Total time of work:
30 minutes for set-up,
4 hours for recording,
4 hours for editing, mixing and mastering.

Supervision and evaluation: Herbert  

We also had to come up with a rough plan of how to do the recording (planning, talking to bands/artists, ...). The general idea was to get a taste of the experience of "studio time is expensive and therefore short", to get used to working under pressure and delivering results with a strictly limited amount of time.

Artist: Roman List (vocals, gittern, bombard, flute, tenor drum)
Song: Neidhart

This is a medieval song. The text tells the story of a party, where it was, who arrived there, what happened at the party, ending with the main narrator, Neidhart, hiding in a barrel of wine so as not to get hit by an angry drunk. Sounds like it was a good party.

You can listen to the recording here. Enjoy!

Roman List performs medieval music in Vienna once in a while, sometimes telling stories about other famous medieval people, such as Count Dracula. It is probably easiest to take a look at his Facebook page for upcoming events.

My next band-related project will probably be mixing recordings for CrashMan's Kapelle, a trio (guitar/vocs, bass, drums) from Vienna playing pop/blues/psychedelic improvisations - in exchange for drum lessons. Looking forward to it!

Sending two XLR out signals to two old loudspeakers

Draft by Gwendolin Korinek, Festspiele Reichenau. CC-BY-SA

This design is a draft which I did not put in practice, but may be useful for theatre purposes.

If you have a pre-installed network cable connection to attach a stagebox (which only provides XLR outs) e. g. on stage left or right, but your FOH is miles away because the theatre hall is quite big, you can save a lot of cable metres with this approach.

The reason you may want to use oldschool loudspeakers is that they could be a functioning prop, part of a play that takes place in, say, the 1970s.

This draft turns two balanced signals into two unbalanced signals.

You need:

- a stagebox 

- two XLR cables

- a special adapter (which you need to solder yourself)

- a 3.5 mm jack cable

- an old household stereo player (which is the „amp“ for the two unbal signals) with a 3.5 mm jack aux input and two oldschool loudspeaker outs

- four loudspeaker cables (2x signals, 2x minus/sleeve)

- two oldschool loudspeakers

- one socket for the stereo player, ideally completely separate from lighting apparatus sockets to avoid hum or issues related to dimming

- stage camouflage to hide the stereo player

Here‘s how it works, in words:

You are sending two signals (double mono, or stereo) to two stagebox XLR outs. Every parameter, including SPL and EQ, can be adjusted for each channel individually, which is very useful if e. g. you want to pre-program a panoramic fade in QLab, or one speaker is in a very different acoustic position from the other.

Next comes the cool adapter you need to solder together as shown in the draft. It uses the classic headphones TRS configuration. One signal to T, the other signal to R, everything else to S.

The two signals now travel to the 3.5 mm jack output which is fed to an old stereo player with a 3.5 mm jack input. Adjust stereo player settings as needed.

The stereo player is the amp for our signals. Next, the signals are fed to two oldschool loudspeaker cables, which feed the two loudspeakers.

For easier use, the stereo player can be hidden in a prop that can be moved around, and only fed with electricity when the play where it is needed actually takes place. Proper cable management (e. g. figure of eight wiring) and formatting cables to the length required (no excess metres of speaker cable) make life easier.

Enjoy!

If you use this design, please credit me (Gwen Korinek) and Festspiele Reichenau.

The 'fake acoustic' synth

This is an idea of a specific synth sound setup for e. g. orchestra ensembles in classical music halls. I had this idea while working at Musikverein (as a stagehand, to avoid any confusion) at night once. I was wondering what it would be like to play a piece for synth and orchestra there someday. Some fresh air, so to say. 

This idea has maybe been around and even put into practice in some keyboard models for a while, I guess, but maybe not in good enough quality... and I haven't really seen it in use in top performing orchestras / chamber music ensembles yet. Could be worthwhile to give it a new shot - with better gear.

Standard settings:
Synth to DI to mixing board to PA, end of story.

'Fake acoustic' synth:

Imagine that a synth player wants to perform a modern classical symphony with an orchestra ensemble in a concert hall with marvellous acoustics.

Rather than wiring the synth to a PA, I think it would make more sense to equip the synth with loudspeakers directly on the instrument in such a way that it mimics e. g. a piano, with sound levels being similar to a classical piano. The sound would come from two loudspeakers, L and R, that are installed at either end of what would be the area where the wing of a classical piano is opened and the strings of the piano can be seen (Fig. 1).

Fig. 1. The 'fake acoustic' synth (doodle). Keys, patch section and pitch bend wheel (or whatever you like, drumpad keys, etc. Here, it is just plain chaos, because I couldn't be bothered to draw it realistically. It is just a generic representation. It could also look totally different, too). In the area where a classical piano would have a wing and strings, L and R panning. Signal flow: from keys/controls to a (very fast) FFT analyser (because you can pitch bend...) to ensure correct panning at the loudspeakers, to speakers. For noise samples, maybe positioning at the predominant frequency makes sense, or the panning parameters for complex samples could already be saved in the database, and then played accordingly.

Each key of the synth would be panned at the correct position (but fretless, of course) so that each frequency that it can produce could be heard at the same position as it would on a classical piano covering the same octave range.

Advantages of this setup:

1. The instrument would fit in well with orchestral pieces of work from previous centuries, where PA systems were not known yet. It would feel more 'natural'.

2. Easy localisation of the instrument by other players and the audience.

3. By emitting sound this way, the instrument could use the naturally good acoustics of good concert halls, like the rest of the orchestra does. For example, the Musikverein in Vienna.

4. It could also work better in a 'chamber music' setting with e. g. theremin (why not, while we're at it) or string instruments.

The quality would, of course, strongly depend on the quality of the installed speakers, which could be limited by speaker size. Probably some sort of studio monitors with soft membranes and neutral sound would be best.

Please note that this would only make sense in rooms and concert halls with decent acoustics.

Addendum:

Another way to realise this would be to patch the signal via a DI to one or two active wedges facing the audience, approximately at keyboard/synth height, to fake an acoustic signal. Since this should not be any louder than other orchestra instruments, the SPL exposure of other musicians should be negligible. If there is enough space on stage, this could give a more realistic impression of the sound's origin than classic PA panning.

As a follow-up thought, I was wondering how to write this sort of thing in a sheet music version. As far as I know, no one has really bothered to come up with a standardised notation yet. It would also require a total absence of "colouring" by various synth models, to reproduce everything the same way every time, over the centuries. That might only be possible to do with a software emulating synths, like puredata. Could be one hell of a nightmare to even get started. BUT. If we want synth music to last, it would be worth the effort. Synth models may go out of production, reproductions may be inaccurate. But if people find a workaround (like just writing down mathematical operations, like in programming code), the sound will stay the same over centuries. Might be worth investing more time into thinking about methods of accurately reproducing synth sounds, by introducing a unified sheet music notation system and maybe software standard, just like MIDI facilitates communication and control.

Left handed piano

 All my piano life, I have been learning new songs mostly in the following "setup" (let's call it setup 1):

Left hand: Chords, arpeggios, small movements

Right hand: Melody, improvisations

Very rarely, however, have I encountered songs written the other way round (let's call it setup 2):

Left hand: Melody, improvisations

Right hand: Chords, arpeggios, small movements

Conclusion: The piano, as it is taught in many places in the world, is a concept by a majority of right-handed people.

However, to obtain, practice and maintain full dexterity in both hands, I would suggest that any and all piano lessons should be structured the following way:

Setup 1 - Setup 2 - repeat (50/50 balance)

(Later perhaps: two melodies, like in canons, etc. - this is just a suggestion for early practice)

Of course, you shouldn't break your arms, but it is perfectly possible to play bass lines as individual melodies with the left hand. In jazz songs, for example, I think it would be easy to switch hands because the accompanying chords of the voicings are very free, anyway, so you can rotate them any way you like.

The result of the current way of teaching piano is:

Most people, at a beginner or intermediate level, are much more agile in their right hands.

Only if they practice switching "chord hand" and "melody hand" will they practice "ambidexterity". I think that, especially for styles like jazz, this should give any piano students more flexibility in playing e.g. impros and increase overall piano "fluidity".

TRS - unbalanced stereo input in mixing boards

 Classic situation:

Spontaneous decision to feed music into a mixing board via a laptop.

In this situation, most people do not have a USB interface at hand to route the signal via two balanced outs to a mixing board.

What happens instead? They use the headphones out, which classically comes in a TRS configuration - unbalanced L and R out.

Once you feed this into a line in of the mixing board, the board combines both channels to mono.

Instead, there should be at least two line ins for a TRS configuration where L and R are recognised as unbalanced signals and routed to the correct channels. (Two line ins because you may want to stream sound from multiple devices, and one line in should be a backup.)

Right now, I have this exact issue working with an Allen & Heath SQ-16. It does not have cinch, so unbalanced tape ins via cinch are not a workaround option here. Maybe cinch to 6.3 mm jack adapters would work, but again, that requires a bunch of adapters. The more adapters, the higher the risk of equipment failure.

Dwarf keyboard for jazz musicians with tiny hands

 I would like to have a keyboard with

- white keys half the size 
and
- black keys 2/3 the size

of a Novation UltraNova keyboard. It should span 5 octaves and have both MIDI and 6.3 mm balanced L+R jack outputs.

Case in point:

E7b9.

With my tiny hands, I cannot play this on an UltraNova without playing at least one unintended extra key and risking hand damage (see picture). In my opinion, this is yet another case of gender bias in product design.

I am the T-Rex of jazz music

To those who claim that this is doable with more practice: It is not really practical, since I do not have the required anatomy! Everything on a keyboard with dimensions for standard male hands and players is easier if you have big hands and becomes very tedious when you have lady hands. On the Ultranova, my hand spans a ninth without cramps, a tenth is only possible with cramps and will not sound good because I cannot play it with ease. However, if I intend to use all five fingers at once, the maximum interval span gets shorter because I need to crook all my fingers to use all five keys. To be limited by design flaws is not amusing.

I have been told that some Casio models have smaller keys. I have yet to find out if any keyboard according to my wishes exists. If not, I would like to see it made, to make jazz music life for children and people with small hands easier.

Off switch for household intercom speakers or doorbells

 All household intercoms should be made with an off switch for the speakers. Alternatively, a potentiometer for the volume would not only allow one to switch off the speakers at any given time, but also adjust the volume to a lower level suitable for cat and sound tech ears.

I manually installed two off switches for my own and a neighbours' intercom. Nowadays, I usually leave it on when I ordered takeaway food or expect mail delivery, but keep it turned off at all other times. I also want to attach a resistor to decrease the volume to a fixed level, since installing a potentiometer would require adjustments on the motherboard, which may make it more difficult to return to the default settings of the intercom, if I would like to let the flat to someone else someday.

Advantages of an off switch:

- Allows you to sleep in peace after a night shift. Usually, post employees will ring the doorbell in the morning hours, hoping that one of the neighbours of the receiver of a package will be at home. The intercom at my flat has also had periods of malfunctioning and/or there have been weirdos ringing doorbells in the middle of the night, repeatedly.

- Cats who are easily scared by foreigners and doorbells, like mine, have a less stressful life. This is also why I would like to install the resistor.

- No disruptions during rehearsals and/or home studio time.

- The sound is not very appealing, so less of it increases one's overall quality of life.

The technical heart of our intercom. A moment of triumph as I detached the loudspeaker from the wiring. All of a sudden, my quality of life had increased significantly.

Intercom off-switch that I installed for my neighbour. Off-switch kindly provided by my workmate Elmar, who is a genius at playing around with various Raspberry Pi smart home applications that he designed by himself - for watering plants, switching off lights etc.