Here’s the situation: we have a song that we want to sing for church on Father’s day. The challenge is that it is a tough song for us to learn. I’ve used some open source tools to try to aid our practice and mastering of this song.
Here are my notes…
I used MuseScore to input the song into a soft copy of the music. I put each part on a separate staff:
Then, I used it to transpose the notes from the key on the sheet music into the key in which we were singing it. This had to match the performance track.
With nothing selected, I selected Notes > Transpose. Since nothing was selected, it offered transpose the whole score.
Generating Individual MIDI files
To help with learning the parts, I wanted a sound track with each part individually.
I highlighted the first Alto measure and the last Bass measure. Then, I hit delete. That deleted everything but the Soprano notes. Then, I did File > Save As and saved it as a MIDI format.
I could have gone to Create > Instruments and deleted the staves from there. That probably would have been easier.
Converting MIDI to Sound
Next, I needed convert the MIDI file to a standard wav file to import into Audacity. Timidity is the application of choice, and I didn’t have it installed. I installed it with:
sudo apt-get install timidity
I tried to convert it to wav with this:
timidity Soprano.mid -Ow -o Soprano.wav
That didn’t work because it didn’t like the instrument. It gave me this output:
No instrument mapped to tone bank 0, program 52 - this instrument will not be heard No pre-resampling cache hit Last 4 MIDI events are ignored
To fix that, I overrode the instrument like this:
timidity Soprano.mid -Ow -o Soprano.wav --force-bank=0 --force-program=0
Still, the tempo was a little fast for practicing, so I added -T 65 to slow it down. The full, final command was:
timidity Soprano.mid -Ow -o Soprano.wav --force-bank=0 --force-program=0 -T 65 timidity Alto.mid -Ow -o Alto.wav --force-bank=0 --force-program=0 -T 65 timidity Tenor.mid -Ow -o Tenor.wav --force-bank=0 --force-program=0 -T 65 timidity Bass.mid -Ow -o Bass.wav --force-bank=0 --force-program=0 -T 65
Combining in Audacity
For some reason, the bottom channel had some static on all of the sound bytes. So, I split apart the channels. I used the Track Dropdown menu > Split Stereo Track.
Then, I clicked the “X” to close the bottom track. That effectively deleted it. Next, I repeated the same two steps in my Alto window. I used Ctrl+A to select the whole track in the Alto window. Then, I used Ctrl + C to copy the track. Back in the Soprano window, I used Tracks > Add New > Audio Track to insert a new track. Finally, I pasted the Alto track that I had copied into that new track.
I repeated that for each step so that I had all four tracks in my “Soprano” window. To help keep them straight, I went to the Track dropdown menu > Name … and set the name to the corresponding musical part.
Then, I adjusted the Gain for the soprano track up and the rest of the tracks down. That emphasized that track. I exported it to a Wav file with File > Export.
I repeated that process for each track to create one wav file for each.
I had to sing and record the voices myself. But, I could use the piano parts as aids for singing. I clicked solo on the track for the part that I was singing. That way I could only hear the one part as I record. When I clicked the record button it playing the track in the background. I added my voice as a new track at the bottom.
Afterwards, I created a special wav file for each part just like I did for the piano parts. Actually, I found that I could double click on the gain setting and provide a numerical value for more precision. I set the emphasized track to 3. I set the other voice parts to -6.
After exporting 4 different files, one for each part, I set the Gain back to 0, and had a little fun. I added a Reverb with Effect > Reverb.
Adjusting the Performance Track
I found that the default/normal ripping software in Ubuntu is Sound Juicer. It’s not installed by default, so I had to install it:
sudo apt-get install sound-juicer
I used Sound Juicer to extract Track 1 (demo) and Track 4 (performance in our key). It ripped them into subfolders under my Music folder in the ogg format. I probably could have changed all that in the preferences.
I went to Audacity in a new window, and I used File > Import > Audio to open the Track 1 that I had ripped. It imported with no problem. Now, I used Effects > Change Pitch to adjust the pitch down 3 half-steps, which puts it in the same key/pitch as the performance track that we plan to use.
After that, I used File > Export to export that to a file.
Now, we need a practice CD to play in the car or during practice. I used Brasero to create a Music CD will all of the different tracks that I put together.
We also wanted decent sheet music. I can sight read somewhat and want something to look at. The original music was on 4 separate staves, and I wanted a regular piano staff.
I began by adding a piano staff with the Create > Instruments menu option:
Next, I selected the whole Alto staff. I clicked on the first measure, held the shift key down, and clicked on the last measure of the Alto Staff. Then, I selected the first measure of the Piano treble staff and pasted it there.
Then, I went to Edit > Voices > Exchange Voice 1 – 2. Now, that my alto notes are in voice two, I repeated the same steps to paste the soprano staff in as voice 1.