-1

I am trying to transcribe a song. In the beginning it seemed to be 100 bpm because the notes aligned nicely when quantizing in Cubase. But then suddenly a few notes came that were 1/3 of a second apart. Is there any way to have 1/3 second in 100bpm (maybe some kind of tuplet?) or do I have to add tempo changes back and forth in these tricky parts?

Emil
  • 719
  • 1
  • 6
  • 16
  • 3
    Are you able to determine what is the tempo, and whether it changes or not, by listening to the music? If not, why? Perhaps what you need is practicing ear training together with some basics with music notation (those two often come together). – user1079505 Mar 04 '22 at 23:29
  • Ah, yeah, as @user1079505 comments, it's far easier to understand music in a not-so-analytical way as in the posing of the question. Unlikely that it's really tempo changes... more likely that it's something like triplets inserted in an otherwise-eighth-notes (or other binary) situation. – paul garrett Mar 04 '22 at 23:39
  • 2
    On the other hand... tempo changes do happen. Please edit to 1) direct us to the song if possible, and 2) establish how comfortable you are with ideas like tempo, beat, and note values, so we can answer at the appropriate level. – Andy Bonner Mar 05 '22 at 01:15
  • I added the name of the piece. I think it is pretty new as in less than or approximately a century so I don't think I can share my transcription but have to keep it private. But I am a newbie when it comes to folk music so not sure. I am kind of comfortable, I just have to look up definition of tuplet sometimes. I kind of know how to read beats as they relate to a bar but I forget which one is supposed to be emphasised. – Emil Mar 05 '22 at 07:04
  • I think the b in bpm is supposed to be counting the note that is in the equality sign of the tempo mark, but which one you choose there is just guesswork to me (make the notes pretty maybe?), so I don't really know difference of 100bpm and 200bpm for example. – Emil Mar 05 '22 at 07:19
  • Transcription questions are off-topic. But just to get you started: this is an easy one, 6/8 at ~70bpm, no tempo changes, no tuplets. – PiedPiper Mar 05 '22 at 10:29
  • Well my question is about whether 0.33333 seconds is possible in 100bpm, not about any specific song. I removed the song now to make it on topic. – Emil Mar 05 '22 at 11:02
  • Seems to be in 6/8, 55 or 110 bpm, depending how it's counted. Can't hear any tuplets. – Tim Mar 05 '22 at 12:03
  • It aligns down to the millisecond with 100bpm in the beginning, so I don't think 55bpm can be right. – Emil Mar 05 '22 at 12:22
  • Why downvotes? It should not be off topic now- it is not about transcribing any particular song. – Emil Mar 05 '22 at 16:18
  • It's unfortunate that you deleted the identifying information. But the important part is that you need to understand about "rubato", and about rhythm. There are many, many genres (including Swedish fiddle music) that either don't care about quantizing perfectly, or actually condemn it (my freshman music school class got a lecture titled "Please Don't Play in Time"). You can't determine a tempo—or even rhythms—simply by aligning to a time grid on a computer. – Andy Bonner Mar 08 '22 at 18:08
  • But bottom line: No, this piece has no significant tempo change; Yes, you should expect it to regularly go out of sync with an artificial metronome or time grid; No, you would not need to reflect slight variations in tempo in a transcription; No, you should not use unusual note values as a way to notate rubato or tempo changes (though a few eccentric composers have done so); and no, "fast notes" don't always mean a tempo change; they can just be shorter rhythmic notes. – Andy Bonner Mar 08 '22 at 18:16
  • It was a pain to import the midi into musescore with weird note lengths so I ended up having to do a lot of postprocessing work guesstimating the notes there. But I finally got the piece on paper so I can practice it with more confidence :-) I am a bit uncertain about the chords though but it doesn't matter that much to me. Unfortunately the low notes seem out of range of my violin so I might have to transpose it if I want to solo play it on the violin... – Emil Mar 08 '22 at 19:20
  • 1
    @Emil One of the coolest things about this piece is the non-standard tuning. The G string is tuned down to a D, an octave below the normal D string. This gives it "growly," rattly quality from its reduced tension. And around 1:10 and following, he does double stops, playing the melody on the (normal) D string while using the detuned string as an open drone, or playing in octaves by putting the same finger on both strings. – Andy Bonner Mar 09 '22 at 15:07

2 Answers2

2

OK, we're not allowed to transcribe this piece for you. But we can suggest how to go about transcribing a piece, and we need something to demonstrate on! Perhaps the piece you mentioned in your original question (still visible in the edit history) will do :-)

Listen to the music. Yes, some notes do align with q=100. But are these the ones you should using for your tempo map? Where are the main beats, the places you'd kick the bass drum if playing the tune? Look at the waveform in Cubase. (Any alignment with the bars & beats grid is pure coincidence.) I'm hearing the main beats where I've marked. 'Dum...dum Dum...dum Dum....dum Dum...dum...'

enter image description here

Do you agree? You should also be able to hear there's a triple subdivision going on - 'One, 2, 3 One, 2, 3...' a waltz rather than a One, 2, 3, 4' march or rock beat. So think 3/4 (or 6/8) rather than 4/4. Now you've got a framework for your transcription. There are tools in Cubase to align a tempo map with the audio, if that's the way you want to proceed. Hint - if the tempo's pretty constant (as this piece is, until a slow-down at the very end) don't try to align fussily, with a tempo event on every beat. Strike an average.

Laurence
  • 92,867
  • 5
  • 62
  • 200
  • Btw, although there a bunch of other guesses in the comments, I'm 95% sure from experience with other Swedish folk music that this counts as a waltz, and should be considered 3/4 rather than any kind of /8. – Andy Bonner Mar 08 '22 at 18:13
  • I tend to agree it's a waltz, and 3/4 would be the conventional notation. It could conceivably be notated in 3/8 though. It's the top number that matters! – Laurence Mar 09 '22 at 14:59
1

As some comments have mentioned, it's hard to guage without more information.

Strictly speaking, it's possible for notes to be any duration apart in any time signature playing any rate of BPM. This is because, while your DAW shows you a nice grid of notes, this is a somewhat artificial construct used to make things easier for a computer. Music is for people though: notes can be played whenever it sounds good. The same goes for pitches - we have standard notes, but a violin or fretless bass can plan between notes just as easily as on them.

Players also bend time intentionally, unintentionally, and endlessly. It's quite unlikely that if this is a live recording the timing is perfectly aligned with your grid the whole way through.

If you'd like to notate it, ask whether it's possible you have some rounding error or have mistaken the time signature. If so, it's quite likely one of the following:

If you're in 6/8 (or similar) and you're referring to a beat as three 8ths:

  • A "beat" is 0.6s second long
  • An 8th would be 0.2 seconds long
  • Three 16ths would be 0.3 seconds long

That's very close to three beats per second, and would imply they are playing a note every three 16ths (See the second bar in the Musink screenshot below). This is a pretty standard syncopation pattern.

If you're in 4/4, a quarter would be 0.6s long and a sixteenth would be 0.3 seconds long, which is pretty close to 1/3rd of a second - certainly within error for recording.

Musink screenshot

Musink Help
  • 11
  • 3