Title: | * * Computer Music, MIDI, and Related Topics * * |
Notice: | Conference has been write-locked. Use new version. |
Moderator: | DYPSS1::SCHAFER |
Created: | Thu Feb 20 1986 |
Last Modified: | Mon Aug 29 1994 |
Last Successful Update: | Fri Jun 06 1997 |
Number of topics: | 2852 |
Total number of notes: | 33157 |
This is pretty fascinating. A guy from USENET has developed a program that supposedly allows your IBM-PC to jam on jazz tunes....I'd like to hear this in action sometime!!! From: [email protected] (James Binkley) Subject: ravel rides again Date: 29 May 90 20:56:55 GMT A number of years ago, I decided to try and create a jazz based improvisation system. As a result I ended up first creating a music programming language that runs on the IBM-pc(mpu 401) called Ravel. Ravel is a C-like interpreted programming language that supports MIDI actions in the language. The main thing is that it is concurrent. It is also modular and supports an import/export mechanism for encapsulation. There is a jazz guitarist and jazz teacher named Jerry Hahn. He has a set of books out now (Mel Bay no less) that contain a rule-based system for jazz improvisation that he came up with over many years of teaching. I felt that it was extremely close to being totally algorithmic (but not quite naturally) and decided to use his rule base and my model. The current jazz improv program basically has that rule set coded up. The structure of the program is that there is a user i/f that has some runtime parameters that can be changed by the user. Basically there are four players (a quartet), drummer bassist lead piano The drummer is fairly primitive and is not using anything I can be proud of in terms of algorithms. The piano player basically plays the chords on the beat (and can jam using a different ruleset that is based on various harmonization principles, but I wouldn't rate it too high, poor to bad). The bass player and the lead both use the Jerry Hahn rule set. The base player uses it to a lesser extent. One of the user i/f parameters you can change is a percent of various time durations for notes; e.g, half, quarter, jazz eighths, that sort of thing. The lead player is a much better improviser than I am, but that may not be saying a whole lot. It could be better; it's not bad though. There is a set of known chords and known scales. It can be added to. It's just kept in arrays as symbolic constants at the moment. A piece is an intro + a chord sequence which is looped on forever. The main thing about the system is that it is useful to me as a vehicle for studying improvisation. I can jam with it. The bass and piano alone are enough to improvise with. The bass player doesn't suffer from lack of variety. On the down side: It could stand much work in the area of theme and variation. The drummer arena could stand some real work. (I need more cpu too...). The piano player is a hack but... It's fun. Ravel as a programming language system is now in the public domain as of May 15, 1990. If you want more information, please send U.S. mail to: Jim Binkley 5814 SW Taylor Portland, Oregon 97221. It's free but free with a catch. There is about 370 pages of documentation at this point + 3 disks of tools + music source. It costs $22.30 to get the manual xeroxed (latest round at Kinko's). I am asking slightly more to cover the rest of the materials. If you get a copy, you can distribute it locally. Again write me for details. Jim Binkley [email protected]
T.R | Title | User | Personal Name | Date | Lines |
---|---|---|---|---|---|
2355.1 | PLEEZE ignore misspellings | DOOLIN::HNELSON | Wed May 30 1990 17:59 | 8 | |
I'm going to send away for this. I've been imagining something like it myself (only just barely, like I've been thinking of remote controllers for turning on multisync monitors from outside x-ray range). I'll cheerfully post a copy to COMMUSICers, if anyone else is interested. (After the first couple, requesters will be referred to earlier requesters.) - Hoyt | |||||
2355.2 | I like new music | MILKWY::JANZEN | Tom 228-5421 FXO/28 | Wed May 30 1990 18:05 | 10 |
This is what I don't like about rule-based composition, and why I like my stocastic program for the Amiga: rule-based programs can produce only known styles; they can't make new music. Everything they play I've heard before. It's boring. Have you heard the one in the Computer Museum? It calculates music overnight in batch and plays it during the day. It is boring IMHO. When you LEARN how to use my stocastic program it makes a wide variety of musical textures; is not limited to any instruments; is not limited to tonal harmony. Tom | |||||
2355.3 | requisite smiley face insertion | SALSA::MOELLER | Endorphin addict | Wed May 30 1990 18:59 | 5 |
Not to be stocastic, but the correct spelling is stocHastic. Impressive technology, whether rule-based or the other one.. karl | |||||
2355.4 | CRAP programmer in U.K. | HUNEY::MACHIN | Thu May 31 1990 05:36 | 5 | |
I'm starting work on an automatic music critique program, Critical Rip And Paste. Then we can get rid of all those junk music papers and save a few trees... Richard. | |||||
2355.5 | RE. new music, etc. | PROSE::DIORIO | Kazoos--the great equalizers | Thu May 31 1990 10:59 | 4 |
Re .2 OK Tom, I'm sold on your program. When will it be available for MS-DOS machines? :-) Mike D | |||||
2355.6 | PRNSYS::LOMICKAJ | Jeffrey A. Lomicka | Thu May 31 1990 12:42 | 9 | |
I've never been impressed by any art form generated from randomness. I like order to the art I appreciate, whether it be visual or aural - so have heard neither - I would lean toward the jazz improvisationalist rather than the stochastic generator - particulary since I get the impression that the Jazz player will accept you as one of the players. If I could play lead and let it provide a drummer improvize a bass line and some chord progressions, I think I could have a good time with it. |