mirror of
https://github.com/microtherion/VocalEasel.git
synced 2024-12-22 19:23:59 +00:00
99 lines
3.3 KiB
Plaintext
99 lines
3.3 KiB
Plaintext
|
|
Using MMA files with the Timidity MIDI player.
|
|
|
|
The program timidity is pretty neat. It lets people who don't have a
|
|
real, external sequencer (or a sequencer built in to a sound card) play
|
|
MIDI files though the computer sound system. Most distributions already
|
|
have this program installed. This file attempts to show a few tricks
|
|
I've learned. It is not a timidity how to or primer.
|
|
|
|
Begin by making sure that you have timidity installed. Typing something like
|
|
|
|
timidity <enter>
|
|
|
|
from the command line should bring up a copyright message like:
|
|
|
|
TiMidity++ version 2.13.2 -- MIDI to WAVE converter and player
|
|
Copyright (C) 1999-2004 Masanao Izumo <iz@onicos.co.jp>
|
|
Copyright (C) 1995 Tuukka Toivonen <tt@cgs.fi>
|
|
..............
|
|
|
|
|
|
If you get a message with "command not found" ... then you'll have
|
|
to find and install timidity first. Check the docs and repositories
|
|
for your distribution.
|
|
|
|
Now, try it with a real file:
|
|
|
|
timidity somefile.mid
|
|
|
|
should play the MIDI file specified though your computer sound system.
|
|
If it doesn't, stop and figure it out. Read the manual, etc.
|
|
|
|
Timidity works by using something called a soundfont. Using this it
|
|
converts each MIDI sound instruction into an audio signal. Don't worry
|
|
about how it does that ... but, remember that quality of the generated
|
|
sound relies on the quality for the soundfont. The default distributions
|
|
I've seen have some pretty crappy sounds.
|
|
|
|
So, upgrade the sounds by installing a better font. I've installed something
|
|
called Airfont 340.sf2 and it sounds good. I'll not give a link here (since
|
|
they do move around), but if you Google for "airfont + 340 + sf2" you should
|
|
find it. The file is about 80 MEG, so be patient.
|
|
|
|
To get timidity to use an sf2 file you have to do 2 things:
|
|
|
|
1. Put the file somewhere that timidity can find it. In most cases
|
|
moving the file to /usr/share/timidity should do the trick.
|
|
|
|
2. Update the timidity config file. In my case I had to change 2 files:
|
|
|
|
i. In /usr/share/timidity create the file timidity.cfg. It
|
|
has a single line:
|
|
|
|
soundfont "Airfont 380.sf2" order=0
|
|
|
|
the quotes are there since the file name has a space in it.
|
|
|
|
ii. In /etc/timidity create the link:
|
|
|
|
cd /etc/timidity; ln -s /usr/share/timidity/timidity.cfg
|
|
|
|
Of course, you'll have to do the above as root.
|
|
|
|
|
|
You can use timidity to create a wav file directly from a MMA midi. This
|
|
file can be used in a program like audacity. Simple:
|
|
|
|
timidity -Ow -ooutfile.wav somefile.mid
|
|
|
|
It is fun to split the MIDI tracks into separate audio tracks. That
|
|
way you can change, for example, the volume for the bass part only
|
|
in a program like audacity. Again, timidity is your friend.
|
|
|
|
For this to work, you need to have a dummy sequence point in each
|
|
MMA MIDI track. When you generate the file, use the -0 option:
|
|
|
|
mma -0 somefile.mma
|
|
|
|
will generate somefile.mid with a "tick" at the start of all the tracks.
|
|
Now, check to see what tracks were created:
|
|
|
|
mma -c somefile.mma
|
|
|
|
Assuming that you have MIDI data on tracks 10, 15 and 16:
|
|
|
|
for a in 10 15 16;do timidity -Ow -Q0 -Q-$a -o$a.wav;done
|
|
|
|
A little program, timsplit.py, which does all the above for you
|
|
has been included in the the MMA distribution in the util directory.
|
|
Read the file README.timsplit for details.
|
|
|
|
Have fun,
|
|
|
|
Bvdp, March/2007
|
|
|
|
|
|
|
|
|