Music-Research Digest Sat, 2 Feb 91 Volume 6 : Issue 9 Today's Topics: INFO from the ICTM *** Send contributions to Music-Research@uk.ac.oxford.prg *** Send administrative requests to Music-Research-Request *** Overseas users should reverse UK addresses and give gateway if necessary *** e.g. Music-Research@prg.oxford.ac.uk *** or Music-Research%prg.oxford.ac.uk@nsfnet-relay.ac.uk *** Back issues, index, etc.: send "help" in a message to archive-server *** @uk.ac.oxford.prg (in the UK) or @hplpm.hpl.hp.com (elsewhere) ---------------------------------------------------------------------- Date: Wed, 23 Jan 91 10:39:03 MEZ From: Schaffrath Subject: INFO from the ICTM To: Stephen Page STUDY GROUP ON COMPUTER AIDED RESEARCH in the "INTERNATIONAL COUNCIL for TRADITIONAL MUSIC (ICTM) (Jan. 18, 1991) 1. Summary of the Study Group for Computer Aided research in Ethno- "musicology - Marseilles, 27-30 September, 1990 Emil H. Lubej" (Vienna) presented a new program on IBM for automatic rhythmic transcription of complex signals with short attacks such as drums, xylophone and piano. The program makes it possible to measure frequencies and durations. The results can be written as an ASCII file for further statistical analysis or alternate repre- sentation, as well as in MIDI code for acoustical control. Giuliana Fugazzotto" (Bologna) discussed her analysis of a folk song of- Barcellona Pozzo di Gotto in Sicily using the Spectre program at the Centro di Sonologia Computazionale di Padova. The melographic representation helped to show the existence of a musical micro- bilinguism in the performance of the song. Margot and Johannes Philipp" (Ludwigsburg, Germany) gave a status report on their ISIS (Interactive Signal Inspection System) project. ISIS is an Atari-ST program for sound analysis with a highly interac- tive audio-visual user shell. The shell is prepared to manage em- bedded signal processing modules. Kathryn Vaughn" (UCLA, USA) discussed the use of digitized performances in her work on the tambura drone and for other ethnomusicological research in music cognition. Integration of sound sampling soft- ware and experimental modules for IBM (and soon for MacIntosh) can be implemented so that judgments of categories related to pitch, sound quality, performance context and other relavent musical pa- rameters can be obtained directly from the culture members. Jinmin Zhou" (UMBC, USA) presented an acoustical study of the Qin showing the role of timbre in the playing technique and the relationship between the traditional notation and the sound of the instrument. Andreine and Bernard Bel "(ISTAR France, Marseille) demonstrated and dis- cussed the implications of mapping sound/gestures to emotion and the concept of transformational creative activity with examples in north Indian rhythmics and contemporary dance. Bernard Bel" (ISTAR France - CNRS) also presented a paper on the develop- ment of "a so-called cellular automaton composing so-called Tur- kish music", demonstrating an application of Bol Processor BP2 to the simulation of melodic improvisation techniques. Suvarnalata Rao"'s (SNDT Women's University Bombay) analysis of intona- tion in Hindustani music explored the application of computer ana- lysis of pitch to a qualitative interpretation of shape and ex- pression in the melodic contours of raga performance. Both the graphic output and the numerical data were used to relate the traditional aesthetic principles to the performance parameters. Haruhiro Katayose" (Osaka University, Japan) brought examples of his work on polyphonic pitch tracking for the piano. The innovative use of interpretative processing allows the computer to follow a perfor- mance on the piano and duplicate the natural musical variation every good performer brings to a score. He also presented examples of shamisan transcriptions created from the computer's "listening" to the music. Frederic Boyer "(LMA-CNRS Marseille) showed us some of the research cur- rently in progress which uses wavelet transforms (as opposed to Fourier transforms) to process signals. The method proved success- ful in highlighting transitory phenomenae which correspond to sig- nificant music events. A sounds can be resynthesized from their wavelet transformed, transformations can be easily processed such as changing pitch or the time scale. This could be very useful for the analysis by simulation of complex musical events. The major obstacle right now is the length of time involved in the computa- tion. Peo Oertli" demonstrated his notational system for "Rhythmics" with many examples both in computer-notation and in praxis. He pointed out the major problems encountered in the transcription and the des- cription of African rhythmic systems. Helmut Schaffrath" gave a demonstration of the latest updates of the MAP- PET software-package including new (free) programs to translate ESAC into staff notation and transformations to MIDI-Standard (Sitter), DARMS (Wiering) and SCORE (Nettheim). Jukka Louhivuori" impressed the Study Group by a very user-friendly ar- chiving project on the Macintosh which by the use of "Hypermedia" allows to look at graphics, statistical information, notation, pictures of musicians etc. and to listen to their original music at the same time. Yang Hong" from Beijing, who could not arrive in time, finally handed in her paper on "Analysis and Definition of Kunqu Tone Patterns" which can be forwarded on request. The Study Group is invited by Prof. Elschek to hold the next meeting in Czechoslovakia in late September 1991. At least six members have agreed to read their papers in this meeting. 2 New Literature: Acquisition et Representation de Connaissances en Musique" Ph.D. thesis in Theoretical Computer Science, by Bernard Bel. Defended at Aix-Marseille III University on 28 November 1990 Jury: Jean-Paul ALLOUCHE, Jean-Claude BERTRAND, Eugene CHOURAQUI, Alain GUENOCHE, Otto LASKE, Jean-Claude RISSET, Bernard VECCHIONE. Text in French (208 pages) English Abstract: This study deals with computer representations of musical knowledge bas- ed on two real-scale experiments. The first experiment focusses on know- ledge acquisition in ethnography (Kippen-Bel's project on North Indian drumming): an expert (the musician), ananalyst (the musicologist), and a machine interact in a learning situation. Improvisation schemata through which musicians express their musical ideas are identified and formal- ized with production rules in a formalism derived from generative gram- mars and pattern languages. A deterministic algorithm is introduced for assessing the membership of arbitrary strings to the langage defined by a given (context- sensitive) grammar. A technique for the inductive in- ference of regular languages is presented, enabling automatic knowledge acquisition of syntactic and lexical knowledge. The second experiment is part of the design of a computer environment for musical composition ("Bol Processor BP2"). Here the specific problem centres on time repre- sentation in a discrete structure of "time objects", and the more ge- neral problem on the synchronization of parallel processes. A method is outlined for determining a structure without complete data on the syn- chronization of its objects. The concept of "sound object" is then for- mally introduced. An efficient algorithm is proposed for the time- setting of objects in a structure, given the constraints arising from their metric and topological properties. Keywords: formal languages, membership test, grammatical inference, syn- chronization, time representation. Orders should be sent to: The Librarian, Groupe Representation et Traitement des Connaissances (GRTC) Centre National de la Recherche Scientifique (CNRS) 31 chemin Joseph Aiguier, 13402 Marseille Cedex 9 (France) ...along with an International Money Order (or UNESCO Coupons) for the amount of 180 French Francs in the name of "Groupe Representation et Traitement des Connaissances". Further information: bel@frmop11.bitnet 3 New note processor for 13,000 melodies in Essen With ESTAFF we have developed our own note processor which automati- cally translates any file encoded in ESAC into Staff notation. We hope that its biggest advantage is not that it is free, but that it is com- patible with any of the Essen MAPPET files now numbering some 13.000 documented melodies! As ESAC users will know the melodies are stored in- cluding their phrase structure, which is left untouched after a trans- lation into Staff notation. ESTAFF has several different disk opera- tions, it can be used to write your own notes, it contains a Macro Graphic option to design your own (Pixel) graphics, it automatically translates into Chinese JIANPU notation and automatically places the lyrics neath your staff! It has a graphic option to draw lines, circles etc., a text option containing four different fonts which also can be manipulated and print options which theoretically allow you to program your own (DOT) printers. The music option allows you to listen to the melody on Synthi or the terrible PC speaker. The latest version also allows you to use the Mouse while copying, shifting etc. Any researcher is welcome to order a free copy once s/he has signed the licence agreement included in this news- letter. 4 MAPPET software now translatable Analogous to ESTAFF we are also programming the MAPPET software so that every user can write his own translation in any European language. This may take some time but we have started with ANAlysis since we had to correct a mistake in this program anyway. 4.1 MAPPET: Mistake in ANAlysis and Update ANA 3.1 For unknown reasons there has been a strange mistake in the ANAlysis of forms (for instance pitch form) from Version 1.0 onwards. We have copied the original STAIRS version and changed it so that the new version of ANAlysis now matches the results in B. Jesser's book. At the same time we have programmed some formal changes. For instance the BLANKS in all analysis of forms are substituted by underlines. The latest version should also enable you to translate any text in files and you are welcome to translate them into Italian, Dutch or what- ever as long as you use ASCII-code. If you want to make use of the up- dated version, 4.2 Analysis examples for MAPPET users For the many examples (31 pages) and staff-notation. This exists in German only, but statistics, staff-notation, formal analysis etc. should also be un- derstood by foreign readers. If you seriously want to use ANA, then you might perhaps like to order a copy. If not, remember that this is a "one-man-show" in Essen which will definitely break down under the pres- sure of too many orders. 4.3 Essen data collection now contains 13,000 melodies The number of ESAC songs has grown considerably by some 1200 Luxembourg and Lothring folksongs input by Damien Sagrillo. A comparable file con- taining 63 Irish folksongs is now public domain whereas the Luxembourg etc. files are not yet available (should you have any serious scientific reason to use them, please write to Damien Sagrillo, Wiesenstr. 67, L- 3336 Hellingen, Luxemburg. Sagrillo is not on e-mail). We also expect to put some 2000 Chinese folksongs into the "public domain pool" provided that our Chinese partners agree. ------------------------------ End of Music-Research Digest