jorensix / tarsos Goto Github PK

In the class MidiToWavRenderer when using custom SoundBank to create the wav file their is an attempt to unload the default SoundBank:

this.synth.unloadAllInstruments(soundbank);

soundbank is the custom SoundBank and so will not unload anything. It also means the next attempt to load the custom SoundBank silently fails:

...
this.synth.unloadAllInstruments(soundbank);
final Instrument[] instruments = soundbank.getInstruments();
for (final Instrument instrument : instruments) {
	synth.loadInstrument(instrument);
}
createWavFile(midiFile, outputFile);
...

I am asking as here Tarsos is starting, but audio playback is not possible, no matter which output device I am selecting, something seems to be wrong there or I am not sure. If I use same Tarsos .jar in Virtual Box, in Windows XP, there sound output is working, but there again, some other Tarsos .jar releases are even not starting, so there seem to be differences, and one has to have luck which Tarsos you selected, fitting to your platform or not, in theory Java should work anywhere in same way, via its bytecode. But only in theory, it seems.

My end goal is actually using Tarsos in linux, directly from command line, actually TarsosDSP and your other tool guess_makam, without any gui, if it would work. Thanks for sharing your work.

In the class MidiToWavRenderer the method createWavFile() that allows the user include a custom SoundBank calls the AudioSynthesizer's open() method. It is then calls its' overloaded method which attempts to reopen 'synth':

createWavFile(final Sequence sequence, final File outputFile) {
...
AudioInputStream stream = synth.openStream(format, map);
...
}

This throws an exception and synth is already open:

javax.sound.midi.MidiUnavailableException: Synthesizer is already open

Hi,
in your ReadMe the "executable Tarsos JAR-file" link is incorrect. It should be "http://0110.be/releases/Tarsos/Tarsos-latest/Tarsos-latest.jar".

I wish there was better documentation in TarsosDSP, but anyways, great work!

I'm trying to understand the result of the function "soundPressureLevel".

According to this page: https://en.wikipedia.org/wiki/Sound_pressure

The dbSPL measure starts around 0 (a very quiet room) to >191 (maximum volume).

But according to my tests, the result of function soundPressureLevel goes from -90 to something close to 0.

Could you please explain why we have negative values?

Thanks in advance!