Inputing ms into/as midi?

[sambaji]

Hello,
I want to create a short midi track using absolute onset ms based on the analysis of an online article "Timing and Meter in Mande Drumming from Mali", which explores non-Western non-isosynchronous beat divisions (see attachment). My idea is to then import it into Ableton so I can experiment with it as a groove template. Any ideas on how I would go about this? Is there a midi sequencer where notes can be entered by ms? Or how about converting a text CSV file to MIDI? Thanks.

Like to article: https://mtosmt.org/issues/mto.14.20.1/m ... ondon.html

[antic604]

I want to create a short midi track using absolute onset ms based on the analysis of an online article...

Do you have an audio example of that? If you'd drag it to Ableton and stretched to match a bar or whatever length it is, then you can drag that to the groove pool and it should extract the groove, which will now be a MIDI file basically, that you can apply to any other clip to use that rhythm/groove.

Or you can try straight up converting audio drums to MIDI:
https://www.ableton.com/en/manual/conve ... midi-track

[Uriel Anthony]

Since you mentioned Ableton, if you have and know to use Max for Live, the sequencer object seq reads/writes midi messages in ms and can read from text files.

https://docs.cycling74.com/max8/refpages/seq#read

There's also a crash course on using seq available if you need more info on it:
https://docs.cycling74.com/max8/tutoria ... q=sequence

[sambaji]

Since you mentioned Ableton, if you have and know to use Max for Live, the sequencer object seq reads/writes midi messages in ms and can read from text files.

https://docs.cycling74.com/max8/refpages/seq#read

There's also a crash course on using seq available if you need more info on it:
https://docs.cycling74.com/max8/tutoria ... q=sequence

Good to know. However, I only have the Intro version of Ableton, which doesn't come with MAx for Live.

[sambaji]

I want to create a short midi track using absolute onset ms based on the analysis of an online article...

Do you have an audio example of that? If you'd drag it to Ableton and stretched to match a bar or whatever length it is, then you can drag that to the groove pool and it should extract the groove, which will now be a MIDI file basically, that you can apply to any other clip to use that rhythm/groove.

Or you can try straight up converting audio drums to MIDI:
https://www.ableton.com/en/manual/conve ... midi-track

Thanks for the suggestion. I only have Ableton Intro, which doesn't come with covert audio drums to MIDI. In terms of using the covert to groove, which Ableton Intro 11 does have, I don't have an isolated audio clip to use.

[sambaji]

Csvmidi looks like something that could work although it is not so user-friendly--a command-line utility using the terminal. https://www.fourmilab.ch/webtools/midicsv/ Still open for a GUI option.

[peersh]

Some coding could do the job. A small program in Python could easily read a CSV file, generate MIDI notes from the onset values and write them to a MIDI file for further use. The pretty midi library https://github.com/craffel/pretty-midi takes ms values for onset and duration of notes. It is rather simple to use and works well.
If that sounds too diffcult or cumbersome for a short track, you could compute note values from milliseconds and tempo (BPM) but this would mean kind of quantization. Example: Let BPM be quarter note = 120, then a quarter note = 500 ms, a sixteenth note = 125 ms. Your table shows durations like 121, 215 so you'll have to decide how fine you choose the (western) quantization grid to be into which you squeeze african rhythms.
Good luck and thanks for the interesting article link.

[sambaji]

Some coding could do the job. A small program in Python could easily read a CSV file, generate MIDI notes from the onset values and write them to a MIDI file for further use. The pretty midi library https://github.com/craffel/pretty-midi takes ms values for onset and duration of notes. It is rather simple to use and works well.
If that sounds too diffcult or cumbersome for a short track, you could compute note values from milliseconds and tempo (BPM) but this would mean kind of quantization. Example: Let BPM be quarter note = 120, then a quarter note = 500 ms, a sixteenth note = 125 ms. Your table shows durations like 121, 215 so you'll have to decide how fine you choose the (western) quantization grid to be into which you squeeze african rhythms.
Good luck and thanks for the interesting article link.

Thanks for your input. Although the dataset is small, with only eight bars given, I am currently analyzing the beat divisions in a spreadsheet to see which tuplets the hits most closely line up with. For example, in the 5th and 6th beat at 110bpm, each with four divisions, the first div is on the downbeat, the 2nd and 3rd div is the 2nd and 3rd div of a quintuplet, and the 4th div is the 6th div of septuplet--in other words, the beat divisions are polytuplet (1 of 1, 2 of 5, 3 of 5, 6 of 7). Like jazz, the tuplet distribution likely changes depending upon tempo. This Mali Mande rhythm is technically not polyrhythm as these divisions are not thought as belonging to overlapping rhythms--they are a holistic whole. Because of our reductionistic understanding of rhythms traditionally in the West, the term "polyrhythm" is often reduced to "the simultaneous use of two or more rhythms that are not readily perceived as deriving from one another, or as simple manifestations of the same meter." For a more holistic approach to understanding non-isosynchronous rhythm, check out the below link, "Groove or swing as distributed rhythmic consonance: Introducing the groove matrix"

https://www.researchgate.net/publicatio ... ove_matrix

[jancivil]

I looked into that, very interesting, I'm glad you brought it up.
There is a fairly simple way to put this: what happens is a type of swing that subdivides differently than the usual triplet; ie., long-short, instead of 2:1 is 3:2 or something, 5:3, you-name-it at the lower level. And the researchers seem to have encountered things which aren't simple enough to talk of like that, non-integers or irrationally-weighted 'swing'. So faster tempi would eventually amount to there being a trivial difference vs the usual.

You may want to investigate Malcolm Braff's microrhythm concept.

[fairlyclose]

If i were still an academic i would be analysing this in terms phase locking of dynamical systems. Makes things simpler in some ways. The sadly long dead Jeff Pressing did interesting work in analysing rhythm and was moving much more into dynamic systems when he dies. No diubt others will have followed
https://online.ucpress.edu/mp/article/1 ... tional-and

But for me it is easier to think in terms of phase locking of sensori motor " loops". People feel a particular groove behaviours and its relation to the other performers. This has little or nothing to do with counting but as JeffP pointed out is more about cultural knowledge eg language patterns as expressed through story or even perhaps traversal through landscape