Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Can R+ be somehow configured to perform like Vocal Harmonizer stomp-boxes ? #32

Open
bdutta opened this issue Jul 11, 2021 · 7 comments
Labels
question Further information is requested

Comments

@bdutta
Copy link

bdutta commented Jul 11, 2021

There are some pretty effective busker friendly Vocal Harmonizer stomp-boxes like Zoom V3, Boss VE-2, TC Helicon Perform-VG or TC-Helicon VoiceLive etc, which can do vocal harmony that do typically the following to the signal chain in real-time:

  • Apply compression
  • Apply some light EQ
  • Apply some studio reverb
  • Apply some optional echo
  • Most importantly, do Vocal Harmony by adding another voicing in various configurations, while taking cues from the chords being played on the guitar that sound natural and just-right.
@Stazed
Copy link
Owner

Stazed commented Jul 11, 2021

The first four steps, yes, the most important fifth step, no...

@bdutta
Copy link
Author

bdutta commented Jul 12, 2021

Thanks. Is there any way to extend Rakarrack-plus by adding custom LV2 plugin, or route signal chain through Carla (kxstudio) ? BTW, isn't the Harmonizer effect in Rakarrack suitable to some extent for this ? Can't one configure the key and scale for this Harmonizer -- for instance, from the Effects description HTML doc, the Select Mode of the Harmonizer plugin seems to indicate this being possible:

In this mode the pitch ratio is variale, this value depends of the selected Interval parameter, and the Note
and Chord effect parameters. Rakarrack-plus then recognize the audio note played (Only "melodies" monophonic data can be played in this mode) and modifies the pitch ratio in order to do a musical harmonization with the Tonality/Chord selected in the parameters, of course the twelve tonality's are available and 33 chords:
6,Maj7,lyd,Maj(9),Maj7(9),6/9,+,m,m6,m7,m7(b5),m9,m7(9),m7(11),m(Maj7),m(Maj7) 9),dim,dim7,7,7(Sus4),7(b5),7(9),7(#11),7(13),7(b9),7(b13),7(#9),+Maj7,+7,1+8,1+5,(Sus4),2

@Stazed
Copy link
Owner

Stazed commented Jul 13, 2021

If I understand what you are trying to do, "...do Vocal Harmony by adding another voicing...while taking cues from the chords being played on the guitar"
You also say "...recognize the audio note played (Only "melodies" monophonic data can be played in this mode) and modifies the pitch..."

The R+ MIDI Converter (experimental) can convert monophonic to MIDI. The Harmonizer effects can be controlled by MIDI chord. So you would need to convert the MIDI note to a MIDI chord. You could route the R+ MIDI out to MIDI Chord (X42 plugin suite), then the MIDI Chord back into the R+ Harmonizer.
The R+ MIDI Converter has some embedded latency that may be problematic for live performances.

If the guitar is playing Chords, this would involve having an effect that can convert audio guitar chords to MIDI. The effect would need to be something similar to the R+ MIDI Converter, but would recognize and send MIDI by chord instead of monophonic.

@bdutta
Copy link
Author

bdutta commented Jul 13, 2021

Thanks. That certainly gives me some good ideas to play around with. BTW, I suppose R+ doesn't have any Ubuntu PPA for prebuilt binaries but needs to be built from source, right ? Also, there is no discussion forum or mailinglist for R+ ? Apologies for using Git issues for Q&A.

@Stazed
Copy link
Owner

Stazed commented Jul 13, 2021

Correct, no PPA and no forum, mailing list.
I have no problem with Q&A here.

@Stazed Stazed added the question Further information is requested label Nov 10, 2022
@domichel
Copy link

domichel commented Oct 6, 2023

The Roland guitar synth have a special microphone made of 6 electrically independent coils, one per string, that in order to get 6 monophonic output signals. It is a long time ago, I made a thesis with an existing DSP board with A/D and D/A converters. The work was to plug in my guitar, play melodies and convert its signal into MIDI events. I didn't used a FFT but a very simple arithmetic mean algorithm, that because it is faster to do so.

I begun by measuring the output signal of the guitar with a memory oscilloscope. the algorithm wait for a maximum amplitude of the signal to occur, wait for the following maximum of the same sign to occur, the time between the 2 events gives the note. It was necessary to implement a signal smoothing with a second arithmetic mean on adjacent sampling values, that because when playing very loud, the chords are touching the frets, which make a rebound of the signal which can cause false maximums during the attack of the notes.

@domichel
Copy link

domichel commented Oct 6, 2023

Hopefully, these rebounds are very close to the first maximum. But as the signal during the attack is very sharp, it can be a zero crossing between the first maximum and the rebound, which imply an arithmetic mean on adjacent sampling values is necessary. That number of sampling values is a good candidate for a knob on a GUI or a command line parameter.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
question Further information is requested
Projects
None yet
Development

No branches or pull requests

3 participants