Re: [abcusers] midi2abc (was: Wanted: ABC transcription...)
James == James Allwright [EMAIL PROTECTED] writes: James The cc construct is a problem because it is notated as a James 3:1 ratio but played as a 2:1 ratio. Played by whom? Does the standard document this? -- Laura (mailto:[EMAIL PROTECTED] , http://www.laymusic.org/ ) (617) 661-8097 fax: (801) 365-6574 233 Broadway, Cambridge, MA 02139 To subscribe/unsubscribe, point your browser to: http://www.tullochgorm.com/lists.html
Re: [abcusers] midi2abc (was: Wanted: ABC transcription...)
On Tue 12 Jun 2001 at 06:35AM -0400, Laura Conrad wrote: James == James Allwright [EMAIL PROTECTED] writes: James The cc construct is a problem because it is notated as a James 3:1 ratio but played as a 2:1 ratio. Played by whom? Does the standard document this? Played by abc2midi and played by most players of hornpipes, I believe. If you try changing abc2midi so that it uses a 3:1 ratio, you should find that hornpipes don't sound quite right. Of course, this is all subjective and maybe you play things a bit differently in Boston :-) . As far as I know this is not documented in the standard at all, since it is more musical knowledge than anything to do with abc. James Allwright To subscribe/unsubscribe, point your browser to: http://www.tullochgorm.com/lists.html
Re: [abcusers] midi2abc (was: Wanted: ABC transcription...)
On Tue 12 Jun 2001 at 06:35AM -0400, Laura Conrad wrote: James == James Allwright [EMAIL PROTECTED] writes: James The cc construct is a problem because it is notated as a James 3:1 ratio but played as a 2:1 ratio. Played by whom? Does the standard document this? Played by abc2midi and played by most players of hornpipes, I believe. If you try changing abc2midi so that it uses a 3:1 ratio, you should find that hornpipes don't sound quite right. Of course, this is all subjective and maybe you play things a bit differently in Boston :-) . As far as I know this is not documented in the standard at all, since it is more musical knowledge than anything to do with abc. Hornpipes are usually transcribed straight, rather than in a dotted rhythm. If it's written with lots of s it's more likely to be a strathspey, which is a much more exaggerated rhythm. Does abc2midi apply a stress program even in the absence of an R: field? If so, how does it know which notes to emphasise? Phil Taylor To subscribe/unsubscribe, point your browser to: http://www.tullochgorm.com/lists.html
Re: [abcusers] midi2abc (was: Wanted: ABC transcription...)
The cc construct is a problem because it is notated as a 3:1 ratio but played as a 2:1 ratio. Played by whom? Does the standard document this? Played by abc2midi and played by most players of hornpipes, I believe. If you try changing abc2midi so that it uses a 3:1 ratio, you should find that hornpipes don't sound quite right. Of course, this is all subjective and maybe you play things a bit differently in Boston :-) . As far as I know this is not documented in the standard at all, since it is more musical knowledge than anything to do with abc. There was nothing to say that the example given was a hornpipe (nor are all hornpipes played that way). This would be a disastrous misinterpretation for almost all other uses of the construct. In a strathspey and in much Baroque music, you interpret as something nearer to ; but any tweaks like that *have* to be under user control. - Jack Campin * 11 Third Street, Newtongrange, Midlothian EH22 4PU, Scotland tel 0131 660 4760 * fax 0870 055 4975 * http://www.purr.demon.co.uk/jack/ food intolerance data recipes, freeware Mac logic fonts, and Scottish music To subscribe/unsubscribe, point your browser to: http://www.tullochgorm.com/lists.html
[abcusers] midi2abc (was: Wanted: ABC transcription...)
John Henckel, Laura Conrad, John Burton and Jon F McNamara has all come up with various good ideas for converting midis to sheet music. I might add Finale myself, a professional music typesetting program far more advanced than (and half as expensive as) Sibelius. Unfortunately, most of those ideas do not adress what actually was the issue here: converting midis to *ABC*. Melody Assistant / Harmony Assistant might be an alternative, but honestly I find the program(s) a bit too expensive, especially since you really don't know what you get before you pay (the free demos available are so limited it's impossible to judge from them). midi2abc seems actually to do a fair job interpreting midi files, it's the ABC part of the program that seems to be the problem. Here are a few exemples taken from my attempts to convert Terpsichore: A2 c2-c/B2-B/A| instead of: A2 c2 B3A| G,2-G,/G,2-G,/A, G,2| instead of: G,2 G,3A, G,2| c2-c/e2-e/c| instead of: c2e e2c| B/cdc/B| instead of: B/c/ dcB| The last one seems to be an midi2abc favourite, popping up all the time. Another problem with midi2abc is that most of the rather impressive selection of transcription options simply don't work. There is however, nothing in midi2abc that some rather simple post-processing can't fix. Doing this manually might or might not take a lot of time (no. 1 in Terpsichore took me more than an hour, while no. 29 didn't need any editing at all), but it shouldn't be too hard to write a perl script or something to do most of it. I can't write a perl script to save my life, though. Frank Nordberg To subscribe/unsubscribe, point your browser to: http://www.tullochgorm.com/lists.html
Re: [abcusers] midi2abc (was: Wanted: ABC transcription...)
On Mon, 11 Jun 2001, Frank Nordberg wrote: I can't write a perl script to save my life, though. Curious thought. save_franks_life.pl -- Richard Robinson The whole plan hinged upon the natural curiosity of potatoes - S. Lem To subscribe/unsubscribe, point your browser to: http://www.tullochgorm.com/lists.html
Re: [abcusers] midi2abc (was: Wanted: ABC transcription...)
Frank == Frank Nordberg [EMAIL PROTECTED] writes: Frank There is however, nothing in midi2abc that some rather Frank simple post-processing can't fix. Doing this manually might Frank or might not take a lot of time (no. 1 in Terpsichore took Frank me more than an hour, while no. 29 didn't need any editing Frank at all), but it shouldn't be too hard to write a perl Frank script or something to do most of it. I can't write a perl Frank script to save my life, though. Wouldn't it be even more straightforward to fix midi2abc? -- Laura (mailto:[EMAIL PROTECTED] , http://www.laymusic.org/ ) (617) 661-8097 fax: (801) 365-6574 233 Broadway, Cambridge, MA 02139 To subscribe/unsubscribe, point your browser to: http://www.tullochgorm.com/lists.html
Re: [abcusers] midi2abc (was: Wanted: ABC transcription...)
James Allwright wrote: My guess is that this is a stylistic change made by whoever created the MIDI files. In this particular case I created the midi files myself with Finale, using every trick I know of to optimize them for midi2abc. Actually, I use midi2abc and abc2midi almaost exclusively to transfer files between Finale and ABC, praying for the day someone comes up with a ABC-ETF converter so I wouldn't have to take the detour through midiland anymore. I've already tried praying and begging for that, but maybe some bribery would work: If someone can come up with a direct (two-way) path between ABC and Finale for Macintosh, I swear I'll have Bach's complete choral harmonisations up on the web in ABC format within a month. Unless you have a very old version of midi2abc, you should be able to retrieve exactly the notes from a computer-generated MIDI file that were put in unless the notes get very short. That might be the problem. There are two independent Mac ports of midi2abc. Both share the same bugs, but they might both be based on outdated source code. If you send me the MIDI files you have, I will have a go at converting them. I'd like to see an abc version of Terpsichore on the web, so I'm willing to spend some time on this project. As I mentioned in another thread, I'm almost finished with the few pieces I have access to, but thanks anyway :-) Frank Nordberg To subscribe/unsubscribe, point your browser to: http://www.tullochgorm.com/lists.html
Re: [abcusers] midi2abc (was: Wanted: ABC transcription...)
Richard Robinson wrote: On Mon, 11 Jun 2001, Frank Nordberg wrote: I can't write a perl script to save my life, though. Curious thought. save_franks_life.pl I might pay a lot for that one ;-) Frank To subscribe/unsubscribe, point your browser to: http://www.tullochgorm.com/lists.html
Re: [abcusers] midi2abc (was: Wanted: ABC transcription...)
James == James Allwright [EMAIL PROTECTED] writes: James On Mon 11 Jun 2001 at 01:47PM +0200, Frank Nordberg wrote: B/cdc/B| instead of: B/c/ dcB| The last one seems to be an midi2abc favourite, popping up all the time. James My guess is that this is a stylistic change made by whoever James created the MIDI files. Unless you have a very old version James of midi2abc, you should be able to retrieve exactly the James notes from a computer-generated MIDI file that were put in James unless the notes get very short. Here's an example of running midi2abc on a MIDI created by abc2midi: X:1 T:VI. Now, O now, I needs must part, C: John Dowland O: From The First Booke of songs or Ayres of foure parts, with Tableture for the Lute T:Cantus M:3/2 L:1/2 N:Original clef, C on first line K:Gmix %%MIDI nobarlines %1 |: B2 A | G2 ^F | E2 G | A2 z | B2 d | c2 B | A2 B w:1.~Now, O now, I needs must part, part- ing though I ab- sent w:While I live I needs must love, love lives not when hope is w:2.~Deare, when I from thee am gone, Gone are all my joyes at w:And al- though your sight I leave, Sight where in my joyes doe w:3.~Deare, If I do not re- turne, Love and I shall die to- w:Part we must though now I die, Die I do to part with %2 A2 z |B2 A |G2 ^F | E2 G | A2 z | B d c c B A | G6 :| w:mourn. Ab- sence can no joy im- part: joy once fled can- not re- turne. w:gone. Now at last de- spaire doth prove, love di- vi- ded lov- eth none. w:once. I loved thee and thee a- lone, In whose love I joy- ed once. w:lie, Till that death doth sence be- reave, Ne- ver shall af- fec- tion die. w:gether. For my ab- sence ne- ver mourne, Whom you might have joy- ed ever: w:you. Him des- paire doth cause to lie, Who both lived and di- eth true. %3 |:c2 c | c2 e | d2 e | d2 z | c c B | A c B | A6 | B2 A | w:Sad de- spair doth drive me hence, this des- paire un- kind- nes sends. If that %4 G2 ^F | E2 G | A2 z | B d c c B A G6 :| w:part- ing bee of- fence, it is shee which then of- fends. % input file cantus.mid % format 1 file 3 tracks X: 1 T: M: 3/2 L: 1/8 Q:1/4=120 K:G % 1 sharps % MIDI Key signature, sharp/flats=0 minor=0 % Time signature=3/2 MIDI-clocks/click=36 32nd-notes/24-MIDI-clocks=8 %C:John Dowland %O:From The First Booke of songs or Ayres of foure parts, with Tableture for the Lute %N:Original clef, C on first line %1 %2 %2 %3 %4 %4 V:1 %VI. Now, O now, I needs must part, %C:John Dowland %O:From The First Booke of songs or Ayres of foure parts, with Tableture for the Lute %Cantus %N:Original clef, C on first line %1 B8A4|G8F4|E8G4|A8z4| B8d4|c8B4|A8B4| %2 A8z4| B8A4|G8F4|E8G4|A8z4| B4d4c4-|c3/2c2-c/2B4A4|G12-|G12| B8A4|G8F4|E8G4|A8z4| B8d4|c8B4|A8B4| %2 A8z4| B8A4|G8F4|E8G4|A8z4| B4d4c4-|c3/2c2-c/2B4A4|G12-|G12| %3 c8c4|c8e4|d8e4|d8z4| c4-c3/2c2-c/2B4|A4c4B4|A12-|A12| B8A4| %4 G8F4|E8G4|A8z4| B4d4c4-|c3/2c2-c/2B4A4|G12-|G12| c8c4|c8e4|d8e4|d8z4| c4-c3/2c2-c/2B4|A4c4B4|A12-|A12| B8A4| %4 G8F4|E8G4|A8z4| B4d4c4-|c3/2c2-c/2B4A4|G12-|G12| %@LENGL %@TVI. Now, O now, I needs must part, %C:John Dowland %O:From The First Booke of songs or Ayres of foure parts, with Tableture for the Lute %N:Original clef, C on first line %1 w:1. Now, O now, I needs must part, part-ing though I ab-sent-2 w: mourn. Ab-sence can no joy im-part: joy once fled can-not re-turne.-2 w: mourn. Ab-sence can no joy im-part: joy once fled can-not re-turne.-3 w: Sad de-spair doth drive me hence, this des-paire un-kind-nes sends. If that-4 w: part-ing bee of-fence, it is shee which then of-fends.-4 w: part-ing bee of-fence, it is shee which then of-fends. It doesn't illustrate Frank's problem, but this one might be related: Original: L:1/2 B d c c B A midi2abc: L:1/8 B4d4c4-|c3/2c2-c/2B4A4 Which should be: B4 d4 c4-|c2 c2 B4 A4 So it looks to me like either midi2abc or abc2midi isn't handling the very well. Incidentally, I just had a report from someone who is using my MIDI files to produce scores that his software frequently reads dotted constructions as triplets, so it may well be an abc2midi bug. The misplacement of the words is another obvious problem with this output file. -- Laura (mailto:[EMAIL PROTECTED] , http://www.laymusic.org/ ) (617) 661-8097 fax: (801) 365-6574 233 Broadway, Cambridge, MA 02139
