Taillard, Pierre-André
Email Address
Birth Date
Project
Organizational Units
Job Title
Last Name
First Name
Name
Search results
Theoretical and experimental study of the role of the reed in clarinet playing, Ph.D.Diss
2018-08-21, Taillard, Pierre-André
Simulation en temps réel de l’impédance d’entrée mesurée ou calculée des instruments à vent
2016-04, Taillard, Pierre-André, Silva, Fabrice, Guillemain, Philippe
La synthèse sonore en temps réel des instruments à vent par modèles physiques demande d’approximer une mesure ou un calcul de l’impédance d’entrée à l’aide d’une batterie de filtres digitaux à relativement bas coût de calcul. Dans l’idéal, la détermination des paramètres des filtres devrait être rapide, sans que cela ne prétérite la qualité de l’approximation dans toute la bande passante utile du filtre, tout en étant robuste en présence de mesures bruitées. De plus, le filtre doit impérativement avoir un comportement passif à toute fréquence, y compris au-delà de la dernière fréquence mesurée. Il apparaît que l’analyse de Prony, ainsi que la méthode LSCE (Least Squares Complex Exponentials) qui en est dérivée, permet d’obtenir une batterie de filtres respectant toutes ces exigences, moyennant quelques aménagements. Les fréquences modales complexes sont approximées dans le domaine temporel, alors que les coefficients modaux complexes sont ajustés par moindres carrés dans le domaine spectral. On introduit des points fictifs au-delà de la dernière fréquence mesurée, de manière que l’approximation par les moindres carrés converge vers une solution assurant la passivité du filtre à toute fréquence. Une méthode de reconstruction de l’impédance d’entrée en basse fréquence complète cette étude. La méthode est validée à l’aide de mesures d’impédance de clarinette, de saxophone et de trompette. Pour la clarinette, par exemple, on peut simuler tous les doigtés avec un filtre numérique ne demandant pas plus de 80 multiplications par pas de temps, pour une fréquence de coupure de 4 kHz et une fréquence d’échantillonnage de 44.1 kHz.
Numerical Computation of the Transfer Functions of an Axisymmetric Duct with the Extended Discrete Singular Convolution Method
2014-07-08, Taillard, Pierre-André, Hélie, Thomas, Bensoam, Joël
This work takes part of the "cagima" project (supported by the ANR) which investigates the defects of the tuning of reed musical instruments as well as their homogeneity of emission and timbre. The goal consists in replacing the traditional approach adopted by instrument makers by a global and rational approach in the design of new instruments /ab initio/ (called "logical instruments"), minimizing some identi ed defects. In this context, an interactive virtual model, predictive and con gurable is proposed. Several approaches are available in the literature but the main di culty is to design digital instruments that are accurate (according to measurements) and that can be implemented in real-time. In this paper, an approach based on the Extended Discrete Singular Convolution method (EDSC) is proposed. The temporal operator (including the fractional derivative term for viscothermal losses) is implemented according to the EDSC formalism. The method allows a fast, straightforward and accurate computation of the transfer functions of an axisymmetric duct with an arbitrary pro le. The computation of the case where the losses are dependent from the diameter causes no noticeable di culty. The results are compared to measurements of a trombone bell.
Influence of the reed flow on the intonation of the clarinet
2012, Taillard, Pierre-André, Dalmont, Jean-Pierre, Guillemain, Philippe
The playing frequency of a reed instrument is generally mainly imposed by the resonator. Nevertheless, numerous other factors have an influence on the playing frequency. Among those, the volume velocity pulsated by the reed must be taken into account in the models as it lowers significantly the playing frequency. At first order, this volume velocity can be taken into account by adding an equivalent volume of about 1.cm3 at the entrance of the pipe. However, the concept of equivalent volume is theoretically only valid for linear vibrations i.e., for the clarinet: the non beating reed regime. In the beating regime, this volume is likely to vary quite a lot with the mouth pressure which can lead to intonation problems. In addition, this volume depends much on the reed opening, but experiments performed with an artificial mouth show that for usual openings (0.3-0.7mm) the intonation remains surprisingly stable. This suggests that the reed and mouthpiece makers know how to control this problem by a well chosen design of the lay. A counter-example is obtained by using a ClaripatchTM especially designed so that intonation is particularly difficult to control.
Modal analysis of the input impedance of wind instruments. Application to the sound synthesis of a clarinet
2018, Taillard, Pierre-André
Comparaison de deux clarinettes séparées par deux cent ans d'évolution: Tentative d'hybridation amusante et instructive entre facture instrumentale, modèles physiques et synthèse sonore
2016-04, Taillard, Pierre-André, Patrick, Sanchez
Cette étude amusante et instructive se propose de vérifier par hybridation d’où proviennent les principales différences de sonorité entre une clarinette Sib Böhm moderne (Buffet Crampon modèle Prestige) et une clarinette historique (copie réalisée par R. Tutz d’après H. Grenser, Dresde, vers 1800). On compare l’impédance d’entrée mesurée pour différents doigtés à l’aide d’un pont d’impédance ainsi que la fonction de transfert permettant de lier le signal de pression mesuré par un bec instrumenté et celui mesuré par un microphone disposé devant l’instrumentiste. De ces mesures, on tire un modèle numérique permettant d’approximer le résonateur et le rayonnement. Les simulations permettent de prédire assez précisément certaines caractéristiques, comme la fréquence de jeu. En comparant les sons externes obtenus en intervertissant les modèles de rayonnement, on constate une différence de sonorité audible mais relativement modeste entre le rayonnement de l’instrument moderne et celui de l’instrument historique. En intervertissant physiquement les excitateurs des 2 instruments, à l’aide d’un baril d’adaptation, on constate une différence de sonorité importante, que les modèles numériques d’excitateurs usuels ne sont pas en mesure de reproduire à l’identique, bien qu’on constate lors des simulations une grande sensibilité pour de subtiles modifications du modèle d’excitateur. Ces expériences montrent que l’excitateur est certainement l’élément dont la modélisation acoustique laisse le plus à désirer.
Statistical Estimation of Mechanical Parameters of Clarinet Reeds Using Experimental and Numerical Approaches
2014, Taillard, Pierre-André, Gross, Michel, Dalmont, Jean-Pierre, Kergomard, Jean, Laloë, Franck
A set of 55 clarinet reeds is observed by holography, collecting 2 series of measurements made under 2 different moisture contents, from which the resonance frequencies of the 15 first modes are deduced. A statistical analysis of the results reveals good correlations, but also significant differences between both series. Within a given series, flexural modes are not strongly correlated. A Principal Component Analysis (PCA) shows that the measurements of each series can be described with 3 factors capturing more than 90% of the variance: the first is linked with transverse modes, the second with flexural modes of high order and the third with the first flexural mode. A forth factor is necessary to take into account the individual sensitivity to moisture content. Numerical 3D simulations are conducted by Finite Element Method, based on a given reed shape and an orthotropic model. A sensitivity analysis revels that, besides the density, the theoretical frequencies depend mainly on 2 parameters: EL and GLT . An approximate analytical formula is proposed to calculate the resonance frequencies as a function of these 2 parameters. The discrepancy between the observed frequencies and those calculated with the analytical formula suggests that the elastic moduli of the measured reeds are frequency dependent. A viscoelastic model is then developed, whose parameters are computed as a linear combination from 4 orthogonal components, using a standard least squares fitting procedure and leading to an objective characterization of the material properties of the cane Arundo donax.
Le jeu et la pédagogie de Liszt dans le Paris des années 1830 et leur rapport avec les traditions pianistiques contemporaines
2018, Torbianelli, Edoardo, Taillard, Pierre-André, Perrin-Goy, Suzanne, Torbianelli, Edoardo
An Analytical Prediction of the Bifurcation Scheme of a Clarinet-Like Instrument: Effects of Resonator Losses
2015, Taillard, Pierre-André, Kergomard, Jean
The understanding of the relationship between excitation parameters and oscillation regimes is a classical topic concerning bowed string instruments. The paper aims to study the case of reed woodwinds and attempts to find consequences on the ease of playing. In the minimum model of clarinet-like instruments, three parameters are considered: i) the mouth pressure, ii) the reed opening at rest, iii) the length of the resonator assumed to be cylindrical. Recently a supplementary parameter was added: the loss parameter of the resonator (using the “Raman model”, that considers resonator losses to be independent of frequency). This allowed explaining the extinction of sound when the mouth pressure becomes very large. The present paper presents an extension of the paper by Dalmont et al. (JASA, 2005), searching for a diagram of oscillation regimes with respect to the reed opening and the loss parameter. An alternative method is used, which allows easier generalization and simplifies the calculation. The emphasis is done on the emergence bifurcation: for very strong losses, it can be inverse, similarly to the extinction one for weak losses. The main part of the calculations are analytical, giving clear dependence of the parameters. An attempt to deduce musical consequences for the player is given.
Scraping technique for clarinet reeds derived from a static bending
2012-04-26, Taillard, Pierre-André, Dalmont, Jean-Pierre
Reed scraping is an art mastered only by a few clarinettists. The published empirical methods do not agree on how to determine where to scrape. In order to understand the problematic more clearly, a purely mechanical and simplified approach is attempted. The reed is modelled by finite elements and a simulation of free static bending shows that the longitudinal flexural stress dominates the problem. By targeting a stress field on different areas of the reed, it is observed that the strain is mainly longitudinal and only slightly lateral, due to the strong anisotropy of the cane (Arundo donax). Thus, a bending moment imposed on the axis of the reed causes almost no translation of the edges of the reed. A local decrease in thickness causes a localized increase in curvature, when the stress field remains constant. With this result, a series of ClaripatchTM was developed to impose a localized decrease in curvature on the mouthpiece’s lay. Using these patches and assuming some hypotheses based on observations, simulations and on the viscoelastic properties of the reed, one can deduce how to scrape it, according to the musical preferences of the clarinettist.