I’m continuing the series of posts on the 30 days of focus.
Today I had quite a frustrating experience. The main underlying reason for my frustration was that I was transcribing on the train, without a reference of the actual guitar sound, and software instruments were sounding different from the video. The direct reason was that the part to transcribe was pretty long and fast: I even could not hear how many notes are in the section, until I really started separating them into tiny sections and listening over and over again.
I started with marking two next phrases in the GaradeBand project with the video. To bring in some order, I called these new phrases BI and BII, and renamed the beginning ones to AI, AII, AIII. Here’s how it looks right now:
Then, I changed the key signature in the Guitar Pro tab, to reflect my knowledge that the composition is actually in D minor, and picked up the transcription.
The first few notes were easy to recognize, although their actual length – not that easy. A part of the issue is that the video had been shot as a solo improvisation, so there was no reference for the beats, thus the recording may lack precision in itself.
Up next, there was a longer sequence of very fast notes. I seemed to recognize the pattern of triplets at first but there seemed to be too many of them for the covered distance, especially as I was limiting my choices by the notes of the D minor scale. I transcribed something for this phrase, somehow transcribed BII and returned back to this one.
During the process, I realized that hearing only the notes I was typing into Guitar Pro was not enough, so I added a software piano track to the GarageBand video project and used “musical typing” to try out various possibilities (without recording). However, the software instrument was not much helpful to distinguish the actual pitches, which seemed to be blending together.
Listening to the whole BI on repeat was not helping much. It seemed that whatever idea I took, it would not fit the music. As the phrase is so fast, I could not even count the number of notes in there, let alone easily recognize them. So I started to focusing on a smaller, tiny subsections, listening to them, counting notes in there, and comparing the overall sound with that typed in the Guitar Pro.
I realized that there were indeed triplets going down. But something was off. In the end, looking at the video and taking the trend into account, I put the chromatic parts in. Now it was looking much better. The sound coming from Guitar Pro, however, had a pretty distinct clash caused by the chromatic notes, which I could not hear in the video. I wrote it off to the specifics of the playback.
Listening to the whole piece, I recognized that the durations of the notes are not fully correct. The notes written in the first bar of the new part are probably somewhat longer, and the faster part of BI might be a bit faster than I wrote it. I will probably return to it later on.
I also could hear that the transcription of BII was off: I had it written starting with D5 instead of B4 flat, and finishing on G4 instead of D4. Basically, I had the left-hand pattern almost correct, but wrote it on 1st and 2nd strings instead of 2nd and 3rd. I fixed that.
Then, I somewhat modified the bars connecting BI and BII, inserting in a pause.
Let’s recap the day 2:
- I listened to the video and marked the two phrases that I would have liked to transcribe today. I’ve also come up with more appropriate names for them.
- I set the correct key signature to remove unnecessary accidentals from the formal musical notation.
- I listened to a looped phrase in the video, gathering possibilities, and then typing the most likely option or checking multiple options with a software instrument in GarageBand.
- I drafted the complex part, moved on to transcribe an easier part, and then returned back to improve the former.
- I started looping tiny simpler regions in the complex part, which helped me to verify the score I had and to realize what I had been missing.
- I listened to the score from the very beginning to the end of the transcribed part to see how the new part fits within the overall frame of the composition, and checked my uncertainties against the video.