Um den vollen Funktionsumfang dieser Webseite zu erfahren, benötigen Sie JavaScript. Hier finden Sie die Anleitung wie Sie JavaScript in Ihrem Browser einschalten.

Transiente Vorgänge bei Einfachrohrblattinstrumenten

Project number:	FWF P28655-N32
Project leader:	Dr. Vasileios Chatziioannou
Research facility:	Institute of Music Acoustics (IWK) University of Music and Performing Arts Vienna
Project Team:	Vasileios Chatziioannou, Alex Hofmann, Alexander Mayer, Montserrat Pàmies-Vilà
Cooperation partners:	Computational Acoustic Modeling Laboratory, McGill University, Montreal Sonic Arts Research Centre, Queen's University Belfast
Date of approval:	05.10.2015
Project start:	01.03.2016
Project end:	29.02.2020
Project data:	mdw repository

Summary

The physics of wind instruments have been extensively studied since the past century. A deep understanding of their function has been obtained through both theoretical formulations and computer-aided simulations. Most of the processes that take place during the generation of musical tones are now well understood and can be sometimes reproduced using physical modelling techniques. Of particular interest is the control of the player over an instrument and how certain actions can be used to manipulate the resulting oscillations. This “Musician-acoustic instrument interaction” has been identified as a hot topic in recent acoustics conferences. It involves several challenges due to the nonlinear phenomena that take place during such interactions, as well as the difficulty to obtain accurate measurements to characterise the player-instrument coupling. Measurements with human players must ensure that a non-intrusive experimental setup is designed, so that players may perform under normal conditions. Measurements with artificial excitation systems can provide useful data for performance analysis, as long as they closely imitate human performance. Data obtained by both types of
measurements may be used to inform and validate physical modelling approaches.

Using a combination of numerical and experimental studies as outlined above, this project helped to increase the understanding of the player actions during articulation in single-reed woodwind instruments. In particular this project aimed:

to develop a time-domain model of a single-reed woodwind instrument that is suitable for capturing transient oscillations, taking the effect of the player’s tongue into account, and to ensure that it is numerically stable, thus allowing on-line access to the model parameters [1, 2];
to develop an inverse modelling algorithm that can be applied to transient signals in order to estimate control parameters and that can resynthesize measured signals obtained under real playing conditions using a non-intrusive, easily reproducible experimental setup [3, 4];
to use inverse modelling and experimental data, in order to extend our understanding of nonlinear phenomena that take place at the excitation mechanism during transients, including reed beating and tongue-reed interaction [2, 5, 6];
to evaluate to what extent measurements under artificial blowing conditions can yield beneficial data regarding transient oscillations and tongue articulation [7];
to refine the developed model for real-time sound synthesis applications, offering accurate resynthesis of transients [8, 9];

Towards these goals, other related aspects have been studied, including the development of sensor-reeds for in vivo and in vitro measurements [10, 11, 12], the development of a control interface for a wind-instrument physical model [13] and the analysis of vocal tract effects in clarinet and saxophone performance [14, 15].

During the course of this study, it has been shown that it is possible to formulate a real-time physical-modelling algorithm that can accurately resynthesise waveforms generated by real and artificial woodwind players, including tone transitions [2, 3, 4]. Therefore, while physics-based sound synthesis attempts were previously mostly validated using qualitative comparisons, a quantitative assessment should be expected in new state-of-the-art physical modelling applications. Furthermore, the construction of an artificial blowing machine (including an artificial tongue) led to experiments that managed to closely resemble sound generation by human players, at least for conventional playing techniques [7]. A manual on the use of the artificial blowing machine has been compiled for future usage [16]. Using data obtained by measurements with both the artificial blowing machine and human players, an inverse model has been developed, in order to extract physical model parameters related to articulation. This was the first such attempt involving transient signals, with previous attempts being limited to steady-state sounds. The above results should be considered in future sound resynthesis attempts, also regarding the possibility to adapt the proposed methodology to other types of musical instruments. In order to study vocal tract effects, in relation to transients, a time-domain analysis has been carried out (based on previously developed frequency-domain techniques). Contrary to previous reports, it has been shown that players make use of their vocal tract not only when performing extraordinary tasks, but also during ordinary playing [15]. This opens up questions regarding instrument performance, especially when player actions are difficult to visualise or quantify.
Finally, during the stability and efficiency analysis of the formulated physical model, a method has been developed in order to precalculate the steps required for an iterative solver to converge [8]. This constitutes an essential tool for numerically solving nonlinear equations, particularly so for real-time applications.

References

[1] S. Schmutzhard, V. Chatziioannou, and A. Hofmann. Parameter optimisation of a viscothermal time-domain model for wind instruments. In Proc. Int. Symp. Musical Acoustics, pages 27--30, Montreal, 2017. https://isma2017.cirmmt.mcgill.ca/proceedings/pdf/ISMA_2017_paper_7.pdf

[2] V. Chatziioannou, S. Schmutzhard, M. Pàmies-Vilà and A. Hofmann. Investigating Clarinet Articulation Using a Physical Model and an Artificial Blowing Machine. Acta Acustica united with Acustica 105 (4), 682-694, 2019. https://doi.org/10.3813/AAA.919348

[3] S. Schmutzhard, M. Pàmies-Vilà, A. Hofmann and V. Chatziioannou. Numerical simulation of transients in single reed woodwind instruments. In Proc. International Congress on Acoustics, 2019.

[4] V. Chatziioannou, S. Schmutzhard, A. Hofmann, and M. Pàmies-Vilà. Inverse modelling of clarinet performance. In Proc. International Congress on Sound and Vibration, 2019.

[5] V. Chatziioannou, A. Hofmann, and M. Pàmies-Vilà. An artificial blowing machine to investigate single-reed woodwind instruments under controlled articulation conditions. In Proc. Meetings on Acoustics, volume 31, page 035003, 2017. http://dx.doi.org/10.1121/2.0000794

[6] M. Pàmies-Vilà, A. Hofmann, and V. Chatziioannou. Analysis of tonguing and blowing actions during clarinet performance. Frontiers in psychology, 9:617, 2018. http://dx.doi.org/10.3389/fpsyg.2018.00617

[7] M. Pàmies-Vilà, A. Hofmann, and V. Chatziioannou. Reproducing tonguing strategies in single-reed woodwinds using an artificial blowing machine. The Journal of the Acoustical Society of America 145 (3), 1677, 2019; best student paper award.

[8] V. Chatziioannou, S. Schmutzhard, and S. Bilbao. On iterative solutions for numerical collision models. In 20th Int. Conf. Digital Audio Effects (DAFx17), Edinburgh, 2017. http://www.dafx17.eca.ed.ac.uk/papers/DAFx17_paper_76.pdf

[9] V. Chatziioannou, A. Hofmann and S. Schmutzhard. A real-time physical model to simulate player control in woodwind instruments. in Proc. International Symposium on Musical Acoustics, 2019.

[10] V. Chatziioannou, A. Hofmann, A. Mayer, and T. Statsenko. Influence of strain-gauge sensors on the vibrational behavior of single reeds. In Proc. Meetings on Acoustics, volume 28, page 035001, 2016

[11] A. Hofmann, V. Chatziioannou, A. Mayer, and H. Hartmann. Development of fibre polymer sensor reeds for saxophone and clarinet. In Proc. New Interfaces for Musical Expression, pages 65--68, Brisbane, 2016. http://www.nime.org/proceedings/2016/nime2016_paper0014.pdf

[12] M. Pamiès-Vilà, A. Hofmann, and V. Chatziioannou. Strain to displacement calibration for single-reeds using a high-speed camera. In Proc. Intern. Symp. Musical Acoustics, pages 5--8, Montreal, 2017. https://isma2017.cirmmt.mcgill.ca/proceedings/pdf/ISMA_2017_paper_8.pdf

[13] A. Hofmann, V. Chatziioannou, S. Schmutzhard, G. Erdoğan and A. Mayer. The Half-Physler: An oscillating real-time interface to a tube resonator model. In Proc. NIME19, 2019.

[14] M. Pàmies-Vilà, G. Scavone, A. Hofmann, and V. Chatziioannou. Investigating vocal tract modifications during saxophone performance. In Proc. Meetings on Acoustics, volume 31, page 035002, 2017; best student paper award. http://dx.doi.org/10.1121/2.0000758

[15] M. Pamies-Vila, A. Hofmann and V. Chatziioannou, Vasileios. The influence of the vocal tract on the attack transients in clarinet playing. Journal of New Music Research 49 (2), 126-135, 2020. https://doi.org/10.1080/09298215.2019.1708412

[16] A. Mayer and M. Pàmies-Vilà. RIAM 2.0 - Reed Instrument Artificial Mouth. A Blowing and Tonguing Device for Artificial Excitation of Single-Reed Woodwind Instruments, IWK Tech Report 1‐2023, 2023. DOI↗

Finanziert vom