Lars Frydén

Generating Musical Performances with Director Musices

Director Musices is a program that transforms notated scores into musical performances. It implements the performance rules emerging from research projects at the Royal Institute of Technology (KTH). Rules in the program model performance aspects such as phrasing, articulation, and intonation, and they operate on performance variables such as tone, inter-onset duration, amplitude, and pitch. By manipulating rule parameters, the user can act as a metaperformer controlling different features of the performance, leaving the technical execution to the computer. Different interpretations of the same piece can easily be obtained. Features of Director Musices include MIDI file input and output, rule palettes, graphical display of all performance variables (along with music notation), and user-defined performance rules. The program is implemented in Common Lisp and is available free as a stand-alone application both for Macintosh and Windows platforms. Further information, including music examples, publications, and the software itself, is located online at http:// www.speech.kth.se/music/performance/.

Musical punctuation on the microlevel: Automatic identification and performance of small melodic units*

In this investigation we use the term musical punctuation for the marking of melodic structure by commas inserted at the boundaries that separate small structural units. Two models are presented th ...

Music from Motion: Sound Level Envelopes of Tones Expressing Human Locomotion

The common association of music with motion was investigated in a direct way. Could the original motion quality of different gaits be transferred to music and be perceived by a listener? Measurements of the ground reaction force by the foot during different gaits were transferred to sound by using the vertical force curve as sound level envelopes for tones played at different tempi. Three listening experiments assessed the motion quality of the resulting stimuli. In the first experiment, where the listeners were asked to freely describe the tones, 25% of answers were direct references to motion; such answers were more frequent at faster tempi. In the second experiment, where the listeners were asked to describe the motion quality, about half of the answers directly related to motion could be classified as belonging to one of the categories of dancing, jumping, running, walking, or stumbling. Most gait patterns were clearly classified as belonging to one of these categories, independent of presentation tempo. In the third experiment, the listeners were asked to rate the stimuli on 24 adjective scales. A factor analysis yielded four factors that could be interpreted as Swift vs. Solemn (factor 1), Graceful vs. Stamping (factor 2), Limping vs. Forceful (factor 3), and Springy (factor 4, no contrasting adjective). The results from the three experiments were consistent and indicated that each tone (corresponding to a particular gait) could clearly be categorised in terms of motion.

Performance Rules for Computer-Controlled Contemporary Keyboard Music

A computer program for synthesis of music performance, originally developed for traditional tonal music by means of an analysis-by-synthesis strategy, is applied to contemporary piano music as well ...

Rules for automated performance of ensemble music

Recently developed parts of a computer program are presented that contain a rule system which automatically converts music scores to musical performance, and which, in a sense, can be regarded as a ...

Threshold and Preference Quantities of Rules for Music Performance

In an analysis- by-synthesis investigation of music performance, rules have been developed that describe when and how expressive deviations are made from the nominal music notation in the score. Two experiments that consider the magnitudes of such deviations are described. In Experiment 1, the musicians9 and nonmusicians9 sensitivities to expressive deviations generated by seven performance rules are compared. The musicians showed a clearly greater sensitivity. In Experiment 2, professional musicians adjusted to their satisfaction the quantity by which six rules affected the performance. For most rules, there was a reasonable agreement between the musicians regarding preference. The preferred quantities seemed close to the threshold of perceptibility.

Rules for Expression in the Performance of Melodies

Starting from a text-to-speech conversion programme (Carlson and Gran- strom, 1975), a note-to-tone conversion programme has been developed (Sundberg and Fryden, 1985). It works with a set of ordered rules affecting the performance of melodies written into the computer. Depending on the musical context, each of these rules manipulates various tone parameters, such as intensity level, fundamental frequency, and duration. In the present study the musical effect of nine rules is tested. Ten melodies were played under several rule-implementation conditions, and musically trained listeners rated the musical quality of each performance. The results support the assumption that the musical quality of performances is improved by applying rules.

Speech and music performance: Parallels and contrasts

Speech and music performance are two important systems for interhuman communication by means of acoustic signals. These signals must be adapted to the human perceptual and cognitive systems. Hence ...

Musical performance. A synthesis‐by‐rule approach

The general purpose of the present paper is to demonstrate some effects on the musical acceptability of a performance of a melody, which may be achieved by means of a synthesis‐by‐rule strategy. Both the structure of a melody and the playing technique of the performer seem important to the music perception and musical performance quality. The synthesis equipment used here consists of a computer controlled synthesizer. The input is the melody written in ordinary notation, and optionally complemented with phrase markers. This is the input material to a text‐to‐speech conversion program [Carlson and Granstrom, in Speech Communication II. edited by G. Fant (Stockholm, 1975) pp. 245‐253] now adapted to for notation‐to‐music conversion. Some of the rules relating the performance to the structure operate on a phase level, and other rules operate on sequences of two or three notes only. As regar‐s production aspects, the influence of a raising of the larynx in singing on the perceived musical quality of the perfo...

Expressive Aspects of Instrumental and Vocal Perfomance

Several music computers can now convert an input note file to a sounding performance. Listening to such performances demonstrates convincingly the significance of the musicians’ contribution to music performance; when the music score is accurately replicated as nominally written, the music sounds dull and nagging. It is the musicians’ contributions that make the performance interesting. In other words, by deviating slightly from what is nominally written in the music score, the musicians add expressivity to the music.