New Releases and the New Picture

A new batch of releases from the last weekend: Drumstick libraries 2.4.0, Drumstick Multiplatform MIDI File Player 1.5.1 and Drumstick MIDI Monitor 1.1.0 are out, and here are the main changes for each one:

Drumstick Libraries v2.4.0:

• implemented ticket #29: RIFF RMID file support. New utility dumprmi, and guiplayer updated.
• implemented ticket #32: missing library version functions in File and RT libs
• implemented ticket #33: versioninfo object for windows libraries
• Enabled by default the internal reverb on macOS DLS Synth
• Avoided hardcoded font family name in vpiano

Drumstick Multiplatform MIDI File Player v1.5.1:

• Fixed GH ticket #6: The dependency target "update_helpfiles" of target "dmidiplayer" does not exist. The pandoc utility is optional again.
• Fixed error checking of DwmGetWindowAttribute() call. This caused a problem in Windows 7 running the "Windows Classic" theme.
• Support for RIFF MIDI files, provided by Drumstick 2.4.0
• Czech manual updated, thanks to Pavel Fric

Drumstick MIDI Monitor v1.1.0

• Removed warnings building with Qt >= 5.15
• MIDI texts/lyrics encoding defaults to Latin1
• New options: Qt Style, Dark mode, and internal icon theme based on breeze
• CMake version >= 3.14
• Added SCM Revision to the about box
• Fixed file info dialog (initial tempo, ...)
• Support for RIFF RMID (.rmi) file format

A common point for all the above is the RIFF MIDI file format support. Citing the LOC:

It is a wrapper format for MIDI data, as first specified by Microsoft, and later extended by MIDI.org (an arm of the MIDI Manufacturers Association) to permit the bundling of both MIDI files and Downloadable Sounds (DLS) files. According to Multimedia Programming Interface and Data Specifications 1.0, August 1991.: "The 'RMID' format consists of a standard MIDI file enclosed in a RIFF chunk. Enclosing the MIDI file in a 'RIFF' chunk allows the file to be consistently identified; for example, an 'INFO' list can be included in the file."

The RIFF format specifications were published in 1991 in Microsoft Windows Multimedia Programmer's Reference, with a brief section devoted to RMID (8-31). The extended specification that documented how both MIDI and downloadable sounds could be bundled in an RMID file (technical note RP-029, "Bundling SMF and DLS data in an 'RMID' File"), was published by the MIDI Manusfacturers Association (MMA) in 2000. The co-authors represent IBM Research, Kurzweil, Microsoft, and Sonic Foundry. In 2001, MMA published the XMF_1_0 specification, which they now prefer to RMID.

The implementation in Drumstick::File offers read-only functionality, and requires the existing SMF class to process the MIDI data portion. The DLS processing is missing, because I've been unable to find a single example of usage, and I suspect that nobody has used it in the real world. Anyway, there is a new utility 'drumstick-dumprmi', which is similar to the other commandline utilities printing the file contents as text in the terminal, with the extra functionality of extract/convert the MIDI data into a standalone SMF file.

Last month was published another utility: wrk2mid v1.0.0 which converts WRK Cakewalk files into SMF. This is something that can be done in Windows using the program Cakewalk by BandLab, which is gratis as a beer, but 'wrk2mid' can be compiled and run on Linux and macOS, so maybe it could be useful for someone. I don't expect to see it on Linux distributions any time soon, though. See: https://wrk2mid.sourceforge.io/

There was a post from 2014 with a diagram of the relationship between the Drumstick libraries v1.0 and applications at the time. Now, all my applications depend on Drumstick v2.x and there are new ones too, so may be this is a good time to paint the new picture...

The new picture

It is a pending subject to renew the repertoire included in dmidiplayer as sample songs. Here is a future one, rendered on macOS (Mojave) using dmidiplayer v1.5.1 with FluidSynth v2.2.3 and GeneralUser Soundfont.

The song is "Negra Sombra" with lyrics by Rosalia de Castro (1837-1885), published in 1880 in her book "Follas Novas", and music by Xoan Montes (1840-1899), a composer from my own birthplace (Lugo). The song was premiered in La Habana, Cuba, in 1892. You may recognize the melody if you like choral music, or because an arrangement by Carlos Nuñez and Luz Casal was included in the soundtrack of the Spanish drama film "The Sea Inside" (2004, Spanish: Mar adentro). This rendition is derived from a score and MIDI sequence by Lalus fecit edition 2013 licensed as CC-by-nc-sa.

Publishing this song at YouTube triggered the algorithm claiming it as a copyrighted work by Sony. I am not monetizing my videos at YouTube so it doesn't matter in practice, but it bothers me a lot so I'm disputing the claim. Google and Sony are taking a work that is in the public domain, and they are appropriating it. That may be seen as criminal conversion, or at least cultural appropriation.

Taking Back From Android

Android is an operating system developed by Google around the Linux kernel. It is not like any other Linux distribution, because not only many common subsystems have been replaced by other components, but also the user interface is radically different based on Java language running into a virtual machine called Dalvik.

An example of subsystem removed from the Linux kernel is the ALSA Sequencer, which is a key piece for MIDI input/output with routing and scheduling that makes Linux comparable in capabilities to Mac OSX for musical applications (for musicians, not whistlers) and years ahead of Microsoft Windows in terms of infrastructure. Android did not offer anything comparable until Android 6 (Marshmallow).

Another subsystem from userspace Linux not included in Android is PulseAudio. Instead, OpenSL ES that can be found on Android for digital audio output and input.

But Android also has some shining components. One of them is Sonivox EAS (originally created by Sonic Network, Inc.) released under the Apache 2 license, and the MIDI Synthesizer used by my VMPK for Android application to produce noise. Funnily enough, it provided some legal fuel to Oracle in its battle against Google, because of some Java binding sources that were included in the AOSP repositories. It is not particularly outstanding in terms of audio quality, but has the ability of providing real time wavetable GM synthesis without using external soundfont files, and consumes very little resources so it may be indicated for Linux projects on small embedded devices. Let's take it to Linux, then!

So the plan is: for the next Drumstick release, there will be a Drumstick-RT backend using Sonivox EAS. The audio output part is yet undecided, but for Linux will probably be PulseAudio. In the same spirit, for Mac OSX there will be a backend leveraging the internal Apple DLS synth. These backends will be available in addition to the current FluidSynth one, which provides very good quality, but uses expensive floating point DSP calculations and requires external soundfont files.

Meanwhile, I've published on GitHub this repository including a port of Sonivox EAS for Linux with ALSA Sequencer MIDI input and PulseAudio output. It also  depends on Qt5 and Drumstick. Enjoy!

Sonivox EAS for Linux and Qt:
https://github.com/pedrolcl/Linux-SonivoxEas

Related Android project:
https://github.com/pedrolcl/android/tree/master/NativeGMSynth

Notes about VMPK 0.6.0 for Desktops

Dear Blog,

It has been a long time, but a week later after the release it is time to discuss some aspects of the latest version of our Virtual MIDI Piano Keyboard for desktop computers.

Where to start? an interesting point is the change of architecture with the replacement of RtMIDI by Drumstick-RT. This library is new and homegrown, part of the Drumstick family which includes Drumstick-File and Drumstick-ALSA as well. The motivation to create it was the difficulty of extending RtMIDI with other drivers different to the ones chosen by his author. This was not a problem in the past, because the RtMIDI sources were always included in the application, and any customization was possible and easy. Now, thanks to the Taliban of Linux distributions forcing the dynamic linking of RtMIDI this is simply not feasible - to the hell with the freedom of Free Software!. Throwing away the ipMIDI backend was not an option. On the other hand, with Drumstick RT is not only possible, but new backends can be compiled separately and installed in the system without recompiling neither VMPK nor Drumstick, because they are in fact plugins. By chance it also has fixed the bug reported (in 2009!) in ticket #15: LinuxSampler did not appear among ALSA connections, because LinuxSampler MIDI port has no flag providing proper characterization and RtMIDI (unlike Drumstick RT) filters out that port.

The replacement of the RtMIDI library with Drumstick-RT was a long time plan, not only for VMPK, but for Drumstick as well, that finally took place now. I hope that this shall be a foundation for features like recording/playback in the future. The only thing that maybe would be missing for some users is the jack-midi interface, but on the other hand Unix users will enjoy native OSS support, and also FluidSynth direct output on all operating systems, meaning  also configurable SoundFonts: a very demanded feature for Windows users.

Another long time request finally implemented is the ability of displaying any number of keys, for instance 88 keys, instead of full octaves, starting with any arbitrary white note (ticket #39), like configuring 25 or 49 keys (depending on which device, laptop or tablet, and screen size you have). Congratulations to all the requesters and sorry for keeping you waiting for this feature so long.

Finally, the migration to Qt5 has happened. This means also replacing a dependency from Xlib to XCB, that hopefully will bring future support for wayland/whatever. The victim has been the keyboard grabbing feature, that was only working on Linux thanks to a now lost X11/Qt4 feature. I hope to bring it back in the future with a multiplatform implementation.

There are now binary packages for 32/64 bit Linux users that shall work on any modern distribution, in the form of installers packaged using the excellent BitRock InstallBuilder. That means including all the required dependencies inside the package, in the same way the libraries are included in the Windows and Mac OS X setup packages. In order to reduce the package weight, superfluous things like Jack support were excluded, because the new FluidSynth backend is intended to provide instant audio out of the box without requiring the users to search, find, ask, learn, install, try and tweak. Something that traditional Linux distributions have failed to do, in despite of their duty of integration and making the life easier to their users. I am pretty sure that many Linux distros will fail to provide VMPK native packages for this release like they did for the 0.5.x  series (see Debian, Ubuntu and Fedora repositories for instance). Prove me wrong, and this kind of binary Linux packages would be deprecated.

SoundFonts want to be free

A "SoundFont" file  (suffix .SF2) is a definition of one or several musical instruments, which can be used with synthesizers (hardware or software) to render, or convert musical notes (eg MIDI files, suffix. MID) into digital sound, which can be used by an audio interface and speakers to play music. Another file format for the same purpose is DownLoadableSounds (suffix. DLS). Both include sound samples that can be entirely synthetic or digitized from real instruments.

General MIDI is a very popular specification, that among other things define a palette of instruments. The instrument #1 is a piano, #41 a violin, #57 trumpet, #74 flute... GM SoundFonts offer 128 instruments arranged in that particular order. GS and XG are extensions of this specification.

Linux needs GM soundfonts that could be distributed together with GPL programs, similar to the need of typographic fonts for text rendering applications. Many Linux distributions incorporate the FluidR3 soundfont in their repositories. It is free and produce good quality sound, but is not small: more than 140 megabytes. MuseScore distributes "TimGM6mb" soundfont, which "only" weights 5.8 megabytes.

There are several software synthesizers using SoundFonts. Well known are FluidSynth and TiMidity++, both with free licenses. One lesser known, but no less interesting, is Gervill. It is part of OpenJDK, and therefore GPLv2 licensed. It is implemented in Java, of course. I am not very fond of Java, and I do not usually use it except for commercial projects when the customer requires their use, but this time I will do it only for fun.

Gervill can use SoundFonts, DLS, or WAV files. A very interesting feature is the so-called "Emergency Soundbank", used when no other external SoundFont is available. Definitions of this SoundFont instruments are fully synthetic and follow the GM standard.

This EmergencySoundbank is a Java class, and does not reside on a file, therefore it can not be used on another synthesizer. However, nothing prevents us from creating a Java program that instantiates the class, and stores the instrument definitions on a disk file. How complicated may this program be? Let's see:

$ cat MakeEmergencySoundfont.java
import com.sun.media.sound.*;
public class MakeEmergencySoundfont {
        public static void main(String[] args) throws Exception {
                SF2Soundbank sf2 = EmergencySoundbank.createSoundbank();
                sf2.save("GervillEmergencySoundbank.sf2");
        }
}


7 lines is not much after all. Let's compile it:

$ javac MakeEmergencySoundfont.java
MakeEmergencySoundfont.java:5: warning: com.sun.media.sound.SF2Soundbank is internal proprietary API and may be removed in a future release
                SF2Soundbank sf2 = EmergencySoundbank.createSoundbank();
                ^
MakeEmergencySoundfont.java:5: warning: com.sun.media.sound.EmergencySoundbank is internal proprietary API and may be removed in a future release
                SF2Soundbank sf2 = EmergencySoundbank.createSoundbank();
                                   ^
2 warnings


We have earned two warnings for the naughty boys, because it is ugly to directly use classes in the namespace "com.sun.media.sound.*", and if the Oracle finds out our prank, he could lock the pantry. Here, take another cookie, Neo...

To compile and use this program you only need OpenJDK6 (runtime and compiler). For older Java versions you can get a "gervill.jar" from the project website.

Running the program produces a SoundFont file:


$ java MakeEmergencySoundfont
$ ls GervillEmergencySoundbank.sf2
-rw------- 1 pedro users 1.8M 2011-09-11 12:50 GervillEmergencySoundbank.sf2


The result weights less than 2 megabytes. Of course the quality of the SoundFont is not high, but it is better than installing nothing at all by default, leaving users wondering what do they need to hear something in their programs that depend on software synthesizers.

It is interesting that a similar technique, also very simple, can be used to produce other SoundFonts based on samples of arbitrary sounds. For more details, see the MakeSoundFont example in Gervill's repository.

Finally, a question remains about what license we can use to distribute the file "GervillEmergencySoundbank.sf2" generated by our program. As a general rule, the output of a GPL program has no restrictions. However, in this case the program output does not come from processing input data, but simply dumps the results of running the algorithms included in the EmergencySoundbank class. To play it safe, we should release it as GPL.