• Welcome to the General Discussion forum for UAD users!

    Please note that this forum is user-run, although we're thrilled to have so much contribution from Drew, Will, and other UA folks!

    Feel free to discuss both UAD and non-UAD related subjects!

    1) Please do not post technical issues here. Please use our UAD Support Forums instead.

    2) Please do not post complaints here. Use the Unrest Forum instead. They have no place in the the General Discussion forum.

    Threads posted in the wrong forum will be moved, so if you don't see your thread here anymore, please look in the correct forum.

    Lastly, please be respectful.

Fast, Accurate, Expressive. Audio to CV with Sonicsmith's Convertor+

scratch17

Venerated Member
Originally shown at NAMM 2018, the Convertor+ is a monophonic audio to control voltage convertor that provides highly
accurate, fast and expressive tracking for any audio source. It works. I have some links to demo videos below.

The unit also contains a synth to provide on board sounds. A model with
a more comprehensive synth (Squaver P1+) is also available. The audio to CV conversion is the same quality on both models.

The Converter+ is available now in the USA at Amazon for $356. The Squaver P1+ is $620.

https://www.amazon.com/dp/B07C4XW149

Either model has the same ability to turn audio into control voltage signals for pitch, envelope, gate and trigger signals.
There are CV outputs for all four signals, so it can control any device with CV pitch and gate inputs.

You can send any kind of audio signal into the Convertor+. So you could control a mono synth with your voice, a flute, or a guitar string, for example.
You can even take any audio track from your DAW and turn it into a synth line. Let that sink in.

But first, watch this demo video:

https://www.youtube.com/watch?v=3d9SP3-GRgA

If you know anything about pitch to MIDI conversion, you will be aware that the lower the pitch of a string, the slower the conversion. It is a fact of physics. The low E string on a guitar is notorious for being the slowest string for a pitch to MIDI converter.

So how is the tracking so fast and accurate with a bass? How is it possible?

This pitch to CV conversion is been done with analog circuits. Does the physics still apply to this type of conversion? Yes it does. So why isn't the bass tracking even slower than the low E string on a guitar? Because the analog circuit that does the conversion is far faster than the currently available digital pitch to MIDI solutions.

And, no. It isn't the MIDI protocol itself that is slowing the conversion. MIDI has nothing to do with the conversion process. The difference has to do with the difference between a slow digital process compared to a fast analog process.

Can you see the irony in this? CV was abandoned by the synth industry when MIDI was adopted. Now we've come full circle.

Here is a demo video with guitar:

https://www.youtube.com/watch?v=dJ4iRzdf1hw

Ignore the sound of the built in synth (which is nice) and concentrate on the tracking accuracy, speed and expressiveness.
Notice that the tracking speed and accuracy is flawless. Check out the highly expressive pitch bending.

At the 0:49 second point two guitarists are playing at the same time. Each has a Convertor+. Both play separate synth lines. Both play fast and bend pitch.
The result is a duo-phonic synth line.

Note that the tracking accuracy, speed and expressiveness is not dependent on the internal synth. You will get the same quality tracking if you use the CV outputs to control another hardware synth.

In effect, the Convertor+ provides the same quality pitch tracking and note on/off speed as a MIDI keyboard controller. But the pitch to CV is analog so the level of innate expressiveness of a stringed instrument, wind instrument or human voice provides far better expressiveness than the mod and/or pitch wheel on a keyboard.

So what about fully polyphonic guitar, or any other n-stringed instrument?

Yes, this solution will work. You just need one Convertor+ per string, an n-string poly pickup and a breakout box.

You can find n-string pickups and a breakout box here:

https://www.cycfi.com/

Cycfi makes pickups for guitars with up to 8 strings. They also have pickups to fit bowed instruments such as violins, etc.

I previously posted about the Cycfi pickups on this forum here:

https://uadforum.com/general-discus...pickups-nexus-2-breakout-box-uad-plugins.html

What about MIDI? If it doesn't slow down the tracking, why doesn't the Convertor+ have MIDI. It isn't available yet. You are limited to synths that have CV pitch and gate inputs for now. That means modular and semi modular synths.

However, Noam Lavi, the creator of these products has emailed me that Sonicsmith has a Eurorack module in development
that will convert CV to MIDI with enough CV inputs to convert 6 channels.

He has not given a timeline for release. But based on his smashing success with the Convertor+ I expect he will be successful with the CV to MIDI converter. The question is, can he make money on it so it is a viable commercial product?

If he is successful with this module, anything with MIDI input will be controllable via a Convertor+.

The implications of this are staggering.

Any instrumentalist or vocalist will be able to control MIDI hardware and MIDI software.

Anyone will be able to write a song if they can carry a tune. They don't even need a good voice. Just sing the part into a mic and convert
to MIDI. Then use a sampler to change the vocal track into the instrument you are writing the line for.

You can even sing a drum part. Forget drum machines. Just use your lips! I find that it is far easier to make percussive sounds with my lips than when tapping a drum pad.

Any instrument will be able to play any other instrument. A sax player will be able to play a flute or an organ. Just convert a sax audio track and use a sampler.

What about chords if you are limited to singing or playing a monophonic instrument? Use your DAW to set up a loop. In Logic the Create Tracks setting will automatically create a new track for each loop cycle. The previous track(s) can be monitored during the looping process. That way you can build up chords with a mono source.

Then convert each track and assign a voice to it.

Speaking of voices, notice I have not been talking much at all about synths. Of course this technology works with synths. But synths are just one category of voices.

I believe that the really important breakthrough here will be the availability of sampler technology to all musicians.

Now I want to suggest a live scenario.

Imagine a live setting with a horn section made of a trumpet, a flugelhorn and a trombone. Each instrument has a mic on it which feeds a Convertor+. Each Convertor+ outputs to the CV to MIDI module. And each sends MIDI data to an instance of Kontakt.

Each horn could layer a reed instrument on top of the sound of their acoustic instruments.

How about possibilities with electric guitar? Assuming full polyphony, the guitar can have 14 strings! Simply add samples of an acoustic guitar up an octave from the real strings' pitch and a bass voice down an octave on the bottom two strings.

Finally, I mentioned that you can use an already recorded audio track as a source for input to a Convertor+. Think of this as an after the fact voice conversion tool. Instead of guitar re-amping, this changes the entire instrument, leaving the timing, articulation and notes intact.

You only need a single Convertor+, and the CV to MIDI module to accomplish this. But what about chords? How do you work with polyphonic material?

You use Melodyne to split the notes of a chord into separate lines, then transfer each to separate tracks. Now everything is is monophonic. So you run through the re-voicing process as many times as needed based on the maximum number of chord voices. Of course, if you wanted to do it in one or two passes, you could use 4 to 6 Convertors.
 

Eric Dahlberg

Purveyor of musical dreams fullfilled.
Very cool! The Roland SE-02 and Behringer Odyssey both accept CV/Gate inputs and I assume both can convert it to MIDI and USB MIDI. Here is an audio example we created with the Behringer Odyssey:

 

Gitaarwerk

Venerated Member
Super interesting. Too long for me to read at the moment. I would love to have a simple, but good guitar to midi or cv conversion.

Also those pickups...man, they look sweet! but, need to put time in them.


Yhea, if you got a roland A01 (i was a lucky one to have it for a 100), you can use it to control a modular setup, accepts different formats,.. :)
 

dnr

Established Member
Last edited:

scratch17

Venerated Member
[MENTION=548]Eric Dahlberg[/MENTION] said:

The Roland SE-02 and Behringer Odyssey both accept CV/Gate inputs and I assume both can convert it to MIDI and USB MIDI.
I have looked at the SE-02 manual, update manual and the MIDI implementation chart. There is no mention of CV to MIDI conversion.

In the above clip of the Behringer, did you determine whether CV in on the Behringer was reflected in MIDI out info?

For example, did you take a CV and gate output from another synth and connect to the Behringer's CV and gate inputs to try to control a soft synth via the Behringer's MIDI output? I doubt that would work. But that would be needed to use the Behringer as a CV to MIDI converter.

Before I started this thread I did a lot of online research looking specifically for ways to convert CV to MIDI.

I found many references to Eurorack modules that converted MIDI to CV. It seems like Eurorack users wanted to have their non modular synths be able to control their modular rigs.

Strangely, I saw only one MIDI to CV converter. That was from Kenton.

I emailed Sonicsmith about the Kenton MIDI Pro, and Noam replied with the following info. The quote below combines 2 emails.

I have tried the Kenton Pro with our products and if MIDI is the only thing you are after, I would recommend waiting.

The maximum "settling time" in the Kenton [CV] to MIDI pro is 50ms and a lot of times this isn't enough for guitar / voice uses. Definitely not bass instruments.

I'm developing a Eurorack module with special MIDI capabilities which will do it much better.
A release date is unknown but would be later this year.

I'm planning the MIDI expander so it'll be able to receive 6 word outputs and make polyphonic MIDI happen but it's only in development stages right now.

I know Cycfi pickups and they look great for what they do.

I will make it possible to join 6 ConVertors together with a single MIDVertor for a polyphonic MIDI out so if you like Cycfi pickup feel free to grab it.

I never found any reference to 'CV to MIDI' in any synth specs or owner's manual with CV inputs and MIDI outputs. My guess is that the designers never expected a CV keyboard in a pre-MIDI rig to be used to control a synth with MIDI input. And no one is expecting CV control to make a comeback.

Also terribly ironic is that MIDI 2.0 will have the increased (32 bit) resolution and timing which will be used to great advantage by CV controllers due to their increased expressiveness compared to standard MIDI controllers.


My Nu2 + Fishman Fusion Guitar Build


[MENTION=14002]gui[/MENTION]taarwerk said:

Also those pickups...man, they look sweet! but, need to put time in them.
I am going to buy a used Squire Bullet Strat to use as a test bed for a monster guitar build I am contemplating with Cycfi and Fishman Fusion pickups. I figure that if I mangle a $100 guitar I can just throw it away. I suspect I might end up throwing two or three away before I get a good routing template for the actual build. I will be using a Fender Elite so I need to get the final build right the first time I use my router.

I am still trying to determine the pickup complement. The guitar will have one Nu2 Multi pickup with two or three Fusion pickups. I might leave the Fusion Open Core Humbucker off if the build gets too complicated. The Cycfi / Lemo multichannel cable can carry up to 15 audio or audio + CV signals. Each Nu2 outputs 6 channels of audio, so I will have 9 outputs left on the cable for the Fusion pickups and CV.

Here is the proposed pickup configuration, starting from the neck:

Guitar Neck -- > Cycfi Nu2 -- Fusion Greg Koch Gristletone (neck pickup) -- Fusion Open Core Humbucker (neck pickup) -- Greg Koch Gristletone (bridge pickup) --> Bridge

Note that while the Nu2 is a humbucker, it is the size of a single coil. So the pickup complement will look like an HHS configuration even though it will actually be HSHS.

The Fusion pickups have multiple voices. There are two on the Fusion Greg Koch single coil pickups and three available on the Open Core humbucker. One of the three voices on the humbucker is actually a single coil setting achieved by turning off a set of the humbucker's coils.

Normally, you can only do one voice at a time on both of the Greg Koch single coils. And normally their outputs are wired together for mono. But since I am going to have the Lemo cable, I am planning on being able to use separate outputs for all three pickups. And I am going to use my Voodoo Labs GCX to do relay switching of pickup voicing and pickup selection. The same principle works to switch amp channels with the GCX. I have an RJM Mastermind GT10 that I use to control the GCX with my foot.

The pickup selection will be relatively easy electrically. It just requires a switch on each pickup's output that turns the output on and off. The voice switching will be more complicated on the Greg Koch pickups because they are integrated into a single housing with a switch to change both pickups to the same voice. Separate voices may not be possible with these two pickups. The Open Core humbucker will be easy though. So worst case I can have a combination of either single coil voice with any of the three humbucker voices. And I will be able to instantly switch to any combination of pickups. Combined, the voice and pickup combinations will provide me with a huge palette of sounds coming from the conventional mono pickups. And I will also have triphonic output with all three set on.

I will need one relay for each pickup (on/off) to do pickup selection. I will need one relay for two voices on each of the Koch single coil pickup if they can be set to separate voices. So if both must use the same voice, I will have only one relay for both. I will need two relays for the open core humbucker because it has three voices. So at most I will use 3 for pickup selection, 2 for the single coils and 2 for the humbucker.

I plan on talking to Fishman about wiring the Greg Koch Gristletone single coils to get the voice switching, pickup selection and separate output I want.

I am content to use the existing volume and tone pots in the Gristletone pickup's electronics. The body of the Gristletone electronics will be routed into the guitar. Only the volume and tone controls will be set above the pick guard, so it won't look like a Tele's control plate.

If I use the Open Core Humbucker, I will add a volume and tone control for it. I will obviously need a volume and tone control for the Cycfi Nu2 on the guitar. That sounds like a really busy layout, but the will be no switches on the guitar because I will be using relays for all of the pickup selections.

I am using my Mackie Onyx 1220 to mix the Nu2's 6 outputs to stereo. I have a TC Electronics G-Major that I can use for effects. The Mackie has individual direct recording outs that I will connect to my UA Apollo audio interface for recording six dry tracks. The Mackie's direct outputs do not go through any of the mixer's channel strip electronics. They simply split the signal at the inputs.

Considering I am using the Greg Koch Gristletone Tele pickups, why the Strat Elite versus the Tele Elite for the final build? Two reasons. First of all, the Strat comes with a whammy bar. To add one to the Tele would cost $350+. Second, the Strat has a larger pick guard. I likely will have to create one anyway once I've determined where the knobs will go, but the Strat won't look too strange with a custom pick guard. So aesthetics, really.
 
Last edited:

Eric Dahlberg

Purveyor of musical dreams fullfilled.
Did you check out Expert Sleepers or MOTU Volta?

Regarding latency, the nice thing is you could track using the Behringer Odyssey as the sound source and then compensate for the latency when using the recorded MIDI with virtual instruments.
 

scratch17

Venerated Member
[MENTION=548]Eric Dahlberg[/MENTION] asked:

Did you check out Expert Sleepers or MOTU Volta?
From the Volta web page describing signal path:

To use Volta, simply instantiate it as an instrument plug-in in your host software.

You then trigger Volta with MIDI notes and a variety of other control sources, and Volta outputs corresponding control voltage (CV) signals to the audio outputs of a DC-coupled audio interface. The resulting CV signals are routed to the CV inputs of your external hardware. Audio output is routed back to Volta.

...a control voltage signal goes out to the external hardware.

Hi Eric. Sadly, that will not work. The reason is found in the last sentence of the above quote. "CV signal goes out [out of the DAW] to the external hardware". I need to get the output of CV pitch and gate from the Convertor+ to go into the DAW to trigger notes on virtual instruments.

Volta takes a MIDI note from the DAW hosting the Volta plugin and converts it to CV for output from a DC coupled audio interface. That is the reverse direction of the conversion that is needed.

The Expert Sleepers hardware does the same thing as the Volta plugin. It converts MIDI to CV, not CV to MIDI.

Regarding latency, the nice thing is you could track using the Behringer Odyssey as the sound source and then compensate for the latency when using the recorded MIDI with virtual instruments.
I cannot find any documentation for the Odyssey on the Behringer web site. Apparently the Odyssey is not yest available. Unless it can accept CV input and internally convert that to MIDI, it cannot be the sound source.

Guitar --> Convertor+ CV pitch and gate out --> Odyssey CV pitch and gate input --> internal conversion to MIDI --> DAW

Even if the Odyssey was able to accomplish the above task, it would only do so in mono.

For fully polyphonic guitar you need one CV pitch and gate output per string, so that each string controls a separate MIDI track which responds to a separate MIDI channel number. So for polyphonic guitar you need six simultaneous CV to MIDI conversions to meet the basic requirement of the use case here.

The 'separate MIDI track / MIDI channel number' requirement is due to pitch bending. When you use a mod wheel on a keyboard controller, all notes you are playing get the same pitch bend information.

That's a no no on a guitar. Guitarists routinely do pitch bend on one note while playing a chord. But they only want to bend that specific note. They don't want the other notes in the chord to bend at all.

So even if you use a 12 voice synth as a sound source, you want each string to be on its own MIDI track, responding only to its exclusive MIDI channel number. That way, pitch bend data for each string is exclusive to that string.

You have to think of the guitar as a parallel set of six keyboards. Kind of like a 6 manual organ. If you are playing manual 5 and you bend the pitch, you don't want that to affect the first manual too.

By the way, I understand that in the MIDI 2.0 spec, the 'all notes get the same pitch bend data' issue is addressed. So that controllers (like the ROLI) which have the ability to separate pitch bend data on a per note basis will be able to do that.

Once the MIDI 2.0 spec is adopted, it might make this use case requirement moot.
 

scratch17

Venerated Member
Eric, I apologize. I was wrong (as usual) about the Expert Sleepers modules. They do have a CV I/O solution for software. The module is called the ES-8. It can send or receive DC coupled CV to any software that can accept CV input.

You can see it here:

Expert Sleepers - ES-8 USB Audio Interface

From the Expert Sleepers web site description of the ES-8:

The ES-8 is a USB 2.0 class-compliant audio interface in a Eurorack module, with DC-coupled inputs and outputs that can be used for both CVs and audio. As such it is directly compatible with all popular computer-based CV generation software (e.g. Silent Way, Max/MSP, Reaktor, CV Toolkit, zMors, Audulus, Bitwig Studio, etc.) as well as audio and CV processing iOS apps.

The ES-8 also has ADAT inputs & outputs. This could be used for example to expand the I/O count via an Expert Sleepers ES-3 and/or ES-6, or to connect a traditional audio interface to add, say, mic preamps or 1/4" line outputs (for example with the Behringer ADA8200).

Without a USB connection, the ES-8 can be used as a standalone ADAT-to-audio converter, providing DC-coupled I/O expansion for another ADAT-equipped audio interface.

An expansion header allows connection of an Expert Sleepers ES-5, to which in turn you can connect Expert Sleepers gate, CV and MIDI expanders.

The ES-8 has a high quality internal clock, and it can optionally sync to the clock from the input ADAT signal.
Specifications

The ES-8's specifications are as follows:

Panel width: 8HP
Module depth: 50mm
Current draw: 198mA on the +12V rail, 52mA on the -12V rail (at 96kHz)
USB socket: Type B
Driver suport: class-compliant USB 2.0 (no driver required for macOS and iOS - Windows drivers provided for Windows 7, 8 & 10)
Channel count: 12 in, 16 out
Supported sample rates: 44.1kHz, 48kHz, 88.2kHz, 96kHz
Audio sample depth: 24 bit
Output connections: 8x 3.5mm TS jacks, DC-coupled
Maximum output voltage: approximately ±10V DC
Analogue inputs: 4x 3.5mm TS jacks, DC-coupled
Input voltage range: approximately ±10V
Digital I/O: 1x ADAT input, 1x ADAT output
Expansion headers: 1 for ES-5, 1 for ES-6
Note that there are 4 DC coupled inputs. However there are input expansion options. You can use an Expert Sleepers ES-6 to add 6 more DC coupled inputs. The ES-8 has an expansion header on the card.

While the ES-8 isn't inexpensive at $475, it represents great value in my use case. Adding an ES-6 ($180) will provide me all of the CV I/O I need to bring 6 channels of CV pitch and gate into my
audio software.

BTW, you don't have to use the ES-8 with the USB connection.

It will operate in standalone mode too. So you can connect it to your Apollo's ADAT I/O and use it as an expansion unit. Remember, it can pass CV and audio through its I/O.

And here is a very important point: you are not actually sending DC coupled audio into the ADAT port of your Apollo. The A/D converter in the ES-8 has turned it into digital audio.
So there is no issue with potential DC offset.

Watch this video for a really great demo of the ES-8 used with an Arturia Minibrute 2S, some hardware modules in a Eurorack, and a host of software modules. The demo also uses a
MOTU 8A audio interface.

https://www.youtube.com/watch?time_continue=1024&v=6mOAgqcM2eI

Here are a few things I noticed watching the video.

Go to 2:32 in the video. The reviewer has already put an instance of a virtual audio interface module into his virtual rack. He uses that instance for the ES-8 module.
He instantiates a second audio interface module and selects the MOTU 8A. What has this to do with anyone who has an Apollo? The answer is that an Apollo can be seen by the Mac, so
it will be seen by the interface virtual module. And just like the MOTU, Apollos have DC coupled outputs. So depending on how many analog outputs you have available on your Apollo(s),
you will also have more CV outputs from your software.

I was blown away by the ease of how setup was between the hardware and software divide using virtual modules and hardware. Moreover, the ease with which these components are combined,
mixed, and matched is mind boggling. As is the power to do sound design.

It is obvious to me that I've been missing a huge component for my studio.

I have Reaktor 6, which by itself is an ever expanding universe. I can't wait to use it with the ES-8 and the free modules from VCV.

You can find them here:

https://vcvrack.com/

I am also going for Softube's Modular. Its Doepfer licensed modules make it a bargain.

Finally, I must point out that if you have a MIDI interface connected to your computer, Reaktor has a utility block that converts CV to MIDI, so you can control MIDI hardware and software with this setup.

As for my guitar use case, I now have a way to output an audio stream from my guitar, convert it to CV with Sonicsmith's Convertor+, and bring that into my audio software with an ES-8.
 

scratch17

Venerated Member
Curse that video! Now of course I need a Eurorack case with power supply. And I had to look at the Minibrute 2S.

Arturia has a great promo on right now.

https://www.arturia.com/minibrute-2s/overview

Buy a Minibrute 2 or 2S bundle for $799, and you get a free Rackbrute 6U plus a Brutebag travel case free. That's a $500 value for an extra $100 over the retail price.

The Rackbrute comes with a power supply module, power and signal distribution hardware, and a tilting leg system that is really cool. It has space for 171 HP width of modules.

Quick, I need a tourniquet for my wallet.
 

scratch17

Venerated Member
Just noticed this note from the video and link to Expert Sleepers description of cable requirements with the ES-8.

In general special cables are recommended when using a DC-coupled audio interface as a source of CVs. Most audio interfaces have balanced outputs (on TRS ('stereo') jacks or XLRs), while synth CV inputs are unbalanced (usually on TS ('mono') jacks or minijacks). If you use a regular stereo or mono jack lead, you'll be shorting out one of the balanced output signals (usually the R (ring) to the S (shield)). While this probably wouldn't be a problem for normal audio use, when outputting the sustained voltages that are useful as CVs you risk damaging the interface hardware.

The usual recommendation is to make up special cables with a TRS jack at one end and a TS at the other, connecting T->T (tip to tip) and S->S (shield to shield) while leaving the R (ring) floating i.e. unconnected. Such cables can be purchased direct from Expert Sleepers - see the accessories page.

An alternative is to wire a TRS jack to two TS jacks, connecting one T->T/S->S and the other R->T/S->S. This then gives you two copies of the output CV, one of them inverted, which is particularly useful from LFOs. Note that this cable configuration is identical to that of a regular "Y" audio insert lead.
 
UAD Bundle Month
Top