Using and controlling soundfonts on GO

[DavidShaw1970]

Is there a way that I can use and control a soundfont on Grandorgue?
Thank you.

[midimusic]

By soundfont I assume you mean .sf2 or similar fonts. If so - then no, GO does not recognise these fonts. To use these fonts you need an organ which works with fluidsynth.
csw900

[DavidShaw1970]

csw900, Thank you for answering. Yes, I do mean .sf2 or similar fonts. And in reading answer, "then no, GO does not recognise these fonts." Again, Thank you. David A. Shaw

…

________________________________ From: midimusic ***@***.***> Sent: Tuesday, February 13, 2024 3:29 AM To: GrandOrgue/grandorgue ***@***.***> Cc: DavidShaw1970 ***@***.***>; Author ***@***.***> Subject: Re: [GrandOrgue/grandorgue] Using and controlling soundfonts on GO (Discussion #1801) By soundfont I assume you mean .sf2 or similar fonts. If so - then no, GO does not recognise these fonts. To use these fonts you need an organ which works with fluidsynth. csw900 — Reply to this email directly, view it on GitHub<#1801 (comment)>, or unsubscribe<https://github.com/notifications/unsubscribe-auth/AWXPIL6R5I6UAB2R5TGUZUDYTMP57AVCNFSM6AAAAABDFGN7FWVHI2DSMVQWIX3LMV43SRDJONRXK43TNFXW4Q3PNVWWK3TUHM4DINJQG42DA>. You are receiving this because you authored the thread.Message ID: ***@***.***>

[hnb2907]

Hi @DavidShaw1970

It's not impossible, but requires some work to convert the contents of the .sf2 to .wav format. I've recently tried this on a linux based system.

1. Install package "gigtools", which contains program "sf2extract". There might be something similar available for Windoze/Mac, sorry I've no idea as I don't have either. ;)
2. Run "sf2extract" on the .sf2 file, which extracts the .wav's You might need to check them in something like "audacity" to ensure they're in a decent bit and sample rate. I also used audacity to convert my .wav's from mono to stereo, added some stereo widening, panned the sound away from centre of left-right, and tinkered with some other stuff to clean them up; this is probably optional for you.
3. For me, extracting the .sf2 gave 2x .wav sample files per octave (C# and G#).
4. Then run "LoopAuditioneer" on the .wav files to add the loop and release markers, if required.
5. Next edit your .odf file to add a new [Ranknnn] and its Pipennn's.
   In my case I used the extracted C# .wav files for "Pipennn=filepath" relating to B C C# D D# E -and- the extracted G# .wav files for pipes relating to F F# G G# A A#.
   To get the pipes to speak at the correct semitone, you also add "PipennnPitchTuning=xxx" where value xxx is +/-100x cents for each semitone correction required. For a bit of additional realism, I slightly randomised the +/-100 for each "Pipe_nnn_", maybe any additional 3 or 4, to deliberately make the tuning less perfect.
   Similarly I also randomised each Pipe's gain so they don't all speak at exactly the same volume. I did this in the "Organ Settings" window, rather than the ODF with "PipennnGain=yyy", but either method should work.

I think this should get you going :) It's the first octave and bit. It shows how to use 3x .wav files and retune them to speak at the correct semitone pitch.

[Rank002]
Name=Diapason
FirstMidiNoteNumber=24
NumberOfLogicalPipes=97
WindchestGroup=003
Percussive=N
Gain=12
HarmonicNumber=8

; C from C#
Pipe001=.\Diapason\025-C#.wav
Pipe001PitchTuning=-98

; C# original sample
Pipe002=.\Diapason\025-C#.wav
Pipe002PitchTuning=0

; D from C#
Pipe003=.\Diapason\025-C#.wav
Pipe003PitchTuning=101

; D# from C#
Pipe004=.\Diapason\025-C#.wav
Pipe004PitchTuning=199

; E from C#
Pipe005=.\Diapason\025-C#.wav
Pipe005PitchTuning=300

; F from G#
Pipe006=.\Diapason\032-G#.wav
Pipe006PitchTuning=-303

; F# from G#
Pipe007=.\Diapason\032-G#.wav
Pipe007PitchTuning=-201

; G from G#
Pipe008=.\Diapason\032-G#.wav
Pipe008PitchTuning=-100

; G# original sample
Pipe009=.\Diapason\032-G#.wav
Pipe009PitchTuning=0

; A from G#
Pipe010=.\Diapason\032-G#.wav
Pipe010PitchTuning=99

; A# from G#
Pipe011=.\Diapason\032-G#.wav
Pipe011PitchTuning=200

; B from C#
Pipe012=.\Diapason\037-C#.wav
Pipe012PitchTuning=-203

; C from C#
Pipe013=.\Diapason\037-C#.wav
Pipe013PitchTuning=-101

; C# original sample
Pipe014=.\Diapason\037-C#.wav
Pipe014PitchTuning=0

; D from C#
Pipe015=.\Diapason\037-C#.wav
Pipe015PitchTuning=102

A caveat: if the .wav samples contain a tremulant, playing 2+ pipes at the same time, the tremulants will oscillate at obviously different speeds, caused by the "PipennnPitchTuning" differences

If you don't have access to "sf2extract", I don't mind extracting them and providing the .wav's for you, as long as there's no copyright restrictions or similar on the .sf2's.
"LoopAuditioneer" seems to be available for linux/windoze/MAC. This was the most labourious part of the process, to get the loops sounding correctly. If you have single-hit percussive instrument samples, you probably don't need to do it.

Enjoy :)
Chris.