As a musical instrument construction material, wood is both musically and aesthetically pleasing. Easy to work and abundant, it has traditionally been the material of choice. Unfortunately, wood is also a very inconsistent material. Due to great environmental and climatic variations, wooden specimens present large variations in their mechanical properties, even within species of a similar region. Surprisingly, an industry based entirely on acoustics has done very little to account for these variations. For this reason, manufactured wooden guitars are acoustically inconsistent. Previous work has shown that varying the dimensions of a guitar soundboard brace is a good method for taking into account variations in the mechanical properties of the wooden soundboard plate. In this thesis, the effects of a scalloped-shaped brace on the natural frequencies of a brace-plate system have been studied and tools have been developed in order to calculate the dimensions of the brace required to account for variations in the mechanical properties of the plate. It has been shown that scalloped braces can be used to modify two natural frequencies of a brace-plate system simultaneously. Furthermore, the most important criteria in modifying any given frequency of a brace-plate system is the mass and stiffness properties of the brace at the antinode of the given frequency’s associated modeshape. Subsequently, designing a brace for desired system natural frequencies, by taking into account the mechanical properties of the wooden plate, is an inverse eigenvalue problem. Since few methods exist for solving the inverse eigenvalue problem of general matrices, a new method based on the generalized Cayley-Hamilton theorem was proposed in the thesis. A further method, based on the determinant of the generalized eigenvalue problem was also presented. Both methods work well, although the determinant method is shown to be more efficient for partially described systems. Finally, experimental results were obtained for the natural frequencies of simply supported wooden plates, with and without a brace, as well as the inverse eigenvalue determinant method. Good correlation was found between theoretical and experimental results.

L’ethnomusicologie est l’étude de la musique en mettant l’accent sur les aspects culturels, sociaux, matériels, cognitifs et/ou biologiques. Ce sujet de thèse, motivé par Pr. Marc Chemillier, ethnomusicologue au laboratoire CAMS-EHESS, traite du développement d’un système automatique de transcription dédié aux répertoires de musique de la cithare marovany de Madagascar. Ces répertoires sont transmis oralement, résultant d’un processus de mémorisation/ transformation de motifs musicaux de base. Ces motifs sont un patrimoine culturel important du pays, et évoluent en permanence sous l’influence d’autres pratiques et genres musicaux. Les études ethnomusicologiques actuelles visent à comprendre l’évolution du répertoire traditionnel, et de préserver ce patrimoine. Pour servir cette cause, notre travail consiste à fournir des outils informatiques d’analyse musicale pour organiser et structurer des enregistrements audio de cet instrument. La transcription automatique de musique consiste à estimer les notes d’un enregistrement à travers les trois attributs : temps de début, hauteur et durée de note. Notre travail sur cette thématique repose sur l’incorporation de connaissances musicales a priori dans les systèmes informatiques. Une première étape de cette thèse fût donc de générer cette connaissance et de la formaliser en vue de cette incorporation. Cette connaissance explorer les caractéristiques multi-modales du signal musical, incluant le timbre, le langage musical et les techniques de jeu. La recherche effectée dans cette thèse se distingue en deux axes : un premier plus appliqué, consistant à développer un système de transcription de musique dédié à la marovany, et un second plus fondamental, consistant à fournir une analyse plus approfondie des contributions de la connaissance dans la transcription automatique de musique. Notre premier axe de recherche requiert une précision de transcription très bonne (c.a.d. une F-measure supérieure à 95 % avec des tolérances d’erreur standardes) pour faire office de supports analytiques dans des études musicologiques. Pour cela, nous utilisons une technologie de captation multicanale appliquée aux instruments à cordes pincées. Les systèmes développés à partir de cette technologie utilisent un capteur par corde, permettant de décomposer un signal polyphonique en une somme de signaux monophoniques respectifs à chaque corde, ce qui simplifie grandement la tâche de transcription. Différents types de capteurs (optiques, piézoélectriques, électromagnétiques) ont été testés. Après expérimentation, les capteurs piézoélectriques, bien qu’invasifs, se sont avérés avoir les meilleurs rapports signal-sur-bruit et séparabilité inter-capteurs. Cette technologie a aussi permis le développement d’une base de données dite “ground truth" (vérité de terrain), indispensable pour l’évaluation quantitative des systèmes de transcription de musique. Notre second axe de recherche propose des investigations plus approfondies concernant l’incorporation de connaissance a priori dans les systèmes automatiques de transcription de musique. Deux méthodes statistiques ont été utilisées comme socle théorique, à savoir le PLCA (Probabilistic Latent Component Analysis) pour l’estimation multi-pitch et le HMM (Hidden Markov Models). / Ethnomusicology is the study of musics around the world that emphasize their cultural, social, material, cognitive and/or biological. This PhD sub- ject, initiated by Pr. Marc CHEMILLIER, ethnomusicolog at the laboratory CAMS-EHESS, deals with the development of an automatic transcription system dedicated to the repertoires of the traditional marovany zither from Madagascar. These repertoires are orally transmitted, resulting from a pro- cess of memorization/transformation of original base musical motives. These motives represent an important culture patrimony, and are evolving contin- ually under the inuences of other musical practices and genres mainly due to globalization. Current ethnomusicological studies aim at understanding the evolution of the traditional repertoire through the transformation of its original base motives, and preserving this patrimony. Our objectives serve this cause by providing computational tools of musical analysis to organize and structure audio recordings of this instrument. Automatic Music Transcription (AMT) consists in automatically estimating the notes in a recording, through three attributes: onset time, duration and pitch. On the long range, AMT systems, with the purpose of retrieving meaningful information from complex audio, could be used in a variety of user scenarios such as searching and organizing music collections with barely any human labor. One common denominator of our diferent approaches to the task of AMT lays in the use of explicit music-related prior knowledge in our computational systems. A step of this PhD thesis was then to develop tools to generate automatically this information. We chose not to restrict ourselves to a speciprior knowledge class, and rather explore the multi-modal characteristics of musical signals, including both timbre (i.e. modeling of the generic \morphological" features of the sound related to the physics of an instrument, e.g. intermodulation, sympathetic resonances, inharmonicity) and musicological (e.g. harmonic transition, playing dynamics, tempo and rhythm) classes. This prior knowledge can then be used in com- putational systems of transcriptions. The research work on AMT performed in this PhD can be divided into a more \applied research" (axis 1), with the development of ready-to-use operational transcription tools meeting the cur- rent needs of ethnomusicologs to get reliable automatic transcriptions, and a more \basic research" (axis 2), providing deeper insight into the functioning of these tools. Our axis of research requires a transcription accuracy high enough 1 (i.e. average F-measure superior to 95 % with standard error tolerances) to provide analytical supports for musicological studies. Despite a large enthusiasm for AMT challenges, and several audio-to-MIDI converters available commercially, perfect polyphonic AMT systems are out of reach of today's al- gorithms. In this PhD, we explore the use of multichannel capturing sensory systems for AMT of several acoustic plucked string instruments, including the following traditional African zithers: the marovany (Madagascar), the Mvet (Cameroun), the N'Goni (Mali). These systems use multiple string- dependent sensors to retrieve discriminatingly some physical features of their vibrations. For the AMT task, such a system has an obvious advantage in this application, as it allows breaking down a polyphonic musical signal into the sum of monophonic signals respective to each string.

Ce travail de thèse propose un transfert de méthodes industrielles et de recherche que sont le prototypage virtuel et la caractérisation mécanique des structures vers l'artisanat d'art : la fabrication des instruments de musique à cordes, plus particulièrement la guitare et le violon.Ce domaine, historiquement éminemment empirique, est étudié ici avec des critères objectifs. Le travail porte principalement sur le comportementmécanique vibratoire de ces instruments de musique. Le comportement mécanique du matériau utilisé (bois de lutherie) est tout d'abord étudié, et une méthode non-destructive proposée et appliquée à l'épicéa et à l'érable. La fiabilité des modèles numériques et ensuite étudiée par des méthodes de vérification et de validation de modèle, habituellement utilisées dans les domaines industriels et de recherche Les modèles développés sont ensuite utilisés en support à l'étude de phénomènes complexes de l'acoustique musicale appliquée au violon et à la guitare. Pour finir, l'utilisation de ces méthodes numériques pour une application concrète en facture instrumentale (prise de décision en milieu incertain, innovation de la géométrie et des matériaux) est proposée. / This work proposes a methodology for the transfer of industrial and research domain methods toward arts and crafts domain. The virtual prototyping and mechanical characterization of materials are used for the stringed musical instruments making domain. The works is adressed to the guitar and violin making. This domain is historically driven by empricism and it is studied here using objective criteria. The main topic is the mechanical vibratory behaviour of musical instruments. Firstly, the mechanical behaviour of the material used (tonewood) is studied. A non-destructive methodology is proposed and applied on spruce and maple species. The reliability of the numerical models is studied by method usually used in industry and research, the verification and validation process. The models developed are used as a support for the study of complex phenomena in musical acoustics. Finally, the utilization of numerical method for a practical application in instrument making (decision support under uncertainties, geometric and material optimization) is proposed. The methodology for a transfer to art and crafts domain is discussed.

In this thesis a method to implement the radiation characteristics of musical instruments in wave field synthesis systems is developed. It is applied and tested in two loudspeaker systems.Because the loudspeaker systems have a comparably low number of loudspeakers the wave field is synthesized at discrete listening positions by solving a linear equation system. Thus, for every constellation of listening and source position all loudspeakers can be used for the synthesis. The calculations are done in spectral domain, denying sound propagation velocity at first. This approach causes artefacts in the loudspeaker signals and synthesis errors in the listening area which are compensated by means of psychoacoustic methods. With these methods the aliasing frequency is determined by the extent of the listening area whereas in other wave field synthesis systems it is determined by the distance of adjacent loudspeakers. Musical instruments are simplified as complex point sources to gain, store and propagate their radiation characteristics. This method is the basis of the newly developed “Radiation Method” which improves the matrix conditioning of the equation system and the precision of the wave field synthesis by implementing the radiation characteristics of the driven loudspeakers. In this work, the “Minimum Energy Method” — originally developed for acoustic holography — is applied for matters of wave field synthesis for the first time. It guarantees a robust solution and creates softer loudspeaker driving signals than the Radiation Method but yields a worse approximation of the wave field beyond the discrete listening positions. Psychoacoustic considerations allow for a successfull wave field synthesis: Integration times of the auditory system determine the spatial dimensions in which the wave field synthesis approach works despite different arrival times and directions of wave fronts. By separating the spectrum into frequency bands of the critical band width, masking effects are utilized to reduce the amount of calculations with hardly audible consequances. By applying the “Precedence Fade”, the precedence effect is used to manipulate the perceived source position and improve the reproduction of initial transients of notes. Based on Auditory Scene Analysis principles, “Fading Based Panning” creates precise phantom source positions between the actual loudspeaker positions. Physical measurements, simulations and listening tests prove evidence for the introduced methods and reveal their precision. Furthermore, results of the listening tests show that the perceived spaciousness of instrumental sound not necessarily goes along with distinctness of localization. The introduced methods are compatible to conventional multi channel audio systems as well as other wave field synthesis applications. / In dieser Arbeit wird eine Methode entwickelt, um die Abstrahlcharakteristik von Musikinstrumenten in Wellenfeldsynthesesystemen zu implementieren. Diese wird in zwei Lautsprechersystemen umgesetzt und getestet. Aufgrund der vergleichsweise geringen Anzahl an Lautsprechern wird das Schallfeld an diskreten Hörpositionen durch Lösung eines linearen Gleichungssystems resynthetisiert. Dadurch können für jede Konstellation aus Quellen- und Hörposition alle Lautsprecher für die Synthese verwendet werden. Hierzu wird zunächst in Frequenzebene, unter Vernachlässigung der Ausbreitungsgeschwindigkeit des Schalls gerechnet. Dieses Vorgehen sorgt für Artefakte im Schallsignal und Synthesefehler im Hörbereich, die durch psychoakustische Methoden kompensiert werden. Im Vergleich zu anderen Wellenfeldsyntheseverfahren wird bei diesem Vorgehen die Aliasingfrequenz durch die Größe des Hörbereichs und nicht durch den Lautsprecherabstand bestimmt. Musikinstrumente werden als komplexe Punktquellen vereinfacht, wodurch die Abstrahlung erfasst, gespeichert und in den Raum propagiert werden kann. Dieses Vorgehen ist auch die Basis der neu entwickelten “Radiation Method”, die durch Einbeziehung der Abstrahlcharakteristik der verwendeten Lautsprecher die Genauigkeit der Wellenfeldsynthese erhöht und die Konditionierung der Propagierungsmatrix des zu lösenden Gleichungssystems verbessert. In dieser Arbeit wird erstmals die für die akustische Holografie entwickelte “Minimum Energy Method” auf Wellenfeldsynthese angewandt. Sie garantiert eine robuste Lösung und erzeugt leisere Lautsprechersignale und somit mehr konstruktive Interferenz, approximiert das Schallfeld jenseits der diskreten Hörpositionen jedoch schlechter als die Radiation Method. Zahlreiche psychoakustische Überlegungen machen die Umsetzung der Wellenfeldsynthese möglich: Integrationszeiten des Gehörs bestimmen die räumlichen Dimensionen in der die Wellenfeldsynthesemethode — trotz der aus verschiedenen Richtungen und zu unterschiedlichen Zeitpunkten ankommenden Wellenfronten — funktioniert. Durch Teilung des Schallsignals in Frequenzbänder der kritischen Bandbreite wird unter Ausnutzung von Maskierungseffekten die Anzahl an nötigen Rechnungen mit kaum hörbaren Konsequenzen reduziert. Mit dem “Precedence Fade” wird der Präzedenzeffekt genutzt, um die wahrgenommene Schallquellenposition zu beeinflussen. Zudem wird dadurch die Reproduktion transienter Einschwingvorgänge verbessert. Auf Grundlage von Auditory Scene Analysis wird “Fading Based Panning” eingeführt, um darüber hinaus eine präzise Schallquellenlokalisation jenseits der Lautsprecherpositionen zu erzielen. Physikalische Messungen, Simulationen und Hörtests weisen nach, dass die neu eingeführten Methoden funktionieren und zeigen ihre Präzision auf. Auch zeigt sich, dass die wahrgenommene Räumlichkeit eines Instrumentenklangs nicht der Lokalisationssicherheit entspricht. Die eingeführten Methoden sind kompatibel mit konventionellen Mehrkanal-Audiosystemen sowie mit anderen Wellenfeldsynthesesystemen.

La viola caipira est un type de guitare brésilienne largement utilisée dans la musique populaire. Elle comprend dix cordes métalliques organisées en cinq paires, accordées à l'unisson ou à l'octave. Le travail de thèse porte sur l'analyse des spécificités des sons musicaux produits par cet instrument, peu étudié dans la littérature.L'analyse des mouvements des cordes pincées au moyen d'une caméra rapide montre l'importance des vibrations par sympathie qui donnent lieu à un halo sonore, constituant une signature perceptive importante. Ces mesures révèlent également l'existence de chocs entre cordes, qui ont des conséquences très clairement audibles. Les mobilités vibratoires au chevalet sont par ailleurs mesurées au moyen de la méthode du fil brisé, simple de mise en oeuvre et peu couteuse dans la mesure où elle évite l'utilisation d'un capteur d'effort. Associée à une analyse modale haute résolution (méthode ESPRIT), ces mesures permettent de déterminer les déformées modales aux points de couplage corde/caisse et donc de caractériser l'instrument.Une modélisation physique, basées une approche modale, est réalisée à des fins de synthèse sonore. Elle prend en compte les mouvements des cordes selon 2 polarisations, les couplages avec la caisse ainsi que les collisions entre cordes. Ce modèle est qualifié de modèle hybride car il combine une approche analytique pour décrire les vibrations des cordes et des données expérimentales décrivant la caisse.Les simulations dans le domaine temporel rendent compte des principales caractéristiques identifiées de la viola caipira. / The viola caipira is a type of Brazilian guitar widely used in popular music. It consists of ten metallic strings arranged in five pairs, tuned in unison or octave. The thesis work focuses on the analysis of the specificities of musical sounds produced by this instrument, which has been little studied in the literature.The analysis of the motions of plucked strings using a high speed camera shows the existance of sympathetic vibrations, which results in a sound halo, constituting an important perceptive feature. These measurements also reveal the existence of shocks between strings, which lead to very clearly audible consequences. Bridges mobilities are also measured using the wire-breaking method, which is simple to use and inexpensive since it does not require the use of a force sensor. Combined with a high-resolution modal analysis (ESPRIT method), these measurements enable to determine the modal shapes at the string/body coupling points and thus to characterize the instrument.A physical modelling, based on a modal approach, is carried out for sound synthesis purposes. It takes into account the strings motions according to 2 polarizations, the couplings with the body and the collisions between strings. This model is called a hybrid model because it combines an analytical approach to describe the vibrations of strings and experimental data describing the body. Simulations in the time domain reveal the main characteristics of the viola caipira.

La diversité des régimes des instruments de la famille des flûtes a été mise en évidence à de nombreuses reprises : régimes statiques, périodiques, ou non périodiques. Cependant, de nombreux aspects de la dynamique de ces instruments demeurent mal compris. Pour les musiciens comme pour les facteurs d'instruments, les transitions entre régimes revêtent une importance particulière : d'une part elles correspondent à des changements de notes, et d'autre part la production d'un régime donné est conditionnée par les paramètres de facture (liés à la fabrication de l'instrument), et de contrôle (ajustés en permanence par l'instrumentiste). On s'attache dans ce document à caractériser les transitions entre régimes dans les flûtes, en lien avec des problématiques de facture et de jeu. Différentes approches sont mises en place. Des approches expérimentales d'une part, avec des mesures sur musicien et sur bouche artificielle. Par ailleurs, un modèle physique de l'instrument - un système dynamique à retard de type neutre - est étudié, par intégration temporelle d'une part, mais également par collocation orthogonale et continuation, donnant ainsi accès aux diagrammes de bifurcations.Croiser les résultats de ces différentes approches permet de mieux appréhender différents phénomènes : hystérésis associée aux changements de régime, ou mécanisme d'apparition des régimes non périodiques. L'influence de paramètres de facture et de contrôle est également étudiée : le rôle majeur de la géométrie interne du canal des flûtes à bec est mis en évidence, et l'influence de la dynamique de la pression dans la bouche du musicien sur les seuils de changement de régimes est caractérisée. / Various studies have highlighted the diversity of regimes in flute-like instruments : static, periodic or non periodic regimes. However, some aspects of their dynamics remain poorly understood. Both for flute players and makers, transitions between regimes are particularly important : on the one hand, they correspond to a change of the note played, and on the other hand, production of a given regime is determined by parameters related to making and to playing of the instrument. In this document, we are interested in caracteristics of regime change in flute-like instruments, in relation with making and playing issues.Different approches are considered. First, experimental methods, with measurement on both musician and an artificial mouth. On the other hand, a physical model of the instrument - a system of delay differential equations of neutral type - is studied, through time-domain integration, and using orthogonal collocation coupled to numerical continuation. This last approach provides access to bifurcation diagrams.Considering results of these different methods, it becomes possible to better understand different experimental phenomena, such as regime change and associated hysteresis, or production mechanisms of non periodic regimes. Influence of different parameters is further studied : the crucial importance of the channel geometry in recorders is highlighted, and the influence of the mouth pressure dynamics on regime change thresholds is analysed.

The bassoon is a conical woodwind instrument blown with a double-reed mouthpiece. The sound is generated by the periodic oscillation of the mouthpiece which excites the air column. The fundamental frequency of this oscillation is determined to a large extent by the resonances of the air column. These can be varied by opening or closing tone-holes. For any given tone hole setting a fine-tuning in pitch is necessary during playing. Musicians adjust the slit opening of the double-reed by pressing their lips against the opposing reed blades. These so-called embouchure corrections are required to tune the pitch, loudness and sound color of single notes. They may be tedious, especially if successive notes require inverse corrections. However, such corrections are essential: Due to the very high frequency sensitivity of the human ear playing in tune is the paramount requirement when playing music. This implies, that embouchure actions provide an important insight into a subjective quality assessment of reed wind instruments from the viewpoint of the musician: An instrument requiring only small corrections will be comfortable to play. Theoretical investigations of the whole system of resonator, reed, and musician by use of a physical model nowadays still seem insufficient with respect to the required precision. Therefore the path of well-described artificial mouth measurements has been chosen here. For the separate treatment of the resonator and the double-reed, existing classical models have been used. Modifications to these models are suggested and verified experimentally. The influence of the musician is incorporated by the lip force-dependent initial reed slit height. For this investigation a measurement setup has been built that allows precise adjustment of lip force during playing. With measurements of the artificial mouth parameters blowing pressure, mouthpiece pressure, volume-flow rate and axial lip position on reed, the experiment is fully described for a given resonator setting represented by an input impedance curve. By use of the suggested empirical model the adjustment parameters can be turned into model parameters. A large data set from blowing experiments covering the full tonal and dynamical range on five modern German bassoons of different make is given and interpreted. The experimental data presented with this work can be a basis for extending the knowledge and understanding of the interaction of instrument, mouthpiece and player. On the one hand, they provide an objective insight into tuning aspects of the studied bassoons. On the other hand the experiments define working points of the coupled system by means of quasi-static model parameters. These may be useful to validate dynamical physical models in further studies. The experimental data provide an important prerequisite for scientific proposals of optimizations of the bassoon and other reed wind instruments. It can further serve as a fundament for the interdisciplinary communication between musicians, musical instrument makers and scientists.

The bassoon is a conical woodwind instrument blown with a double-reed mouthpiece. The sound is generated by the periodic oscillation of the mouthpiece which excites the air column. The fundamental frequency of this oscillation is determined to a large extent by the resonances of the air column. These can be varied by opening or closing tone-holes. For any given tone hole setting a fine-tuning in pitch is necessary during playing. Musicians adjust the slit opening of the double-reed by pressing their lips against the opposing reed blades. These so-called embouchure corrections are required to tune the pitch, loudness and sound color of single notes. They may be tedious, especially if successive notes require inverse corrections. However, such corrections are essential: Due to the very high frequency sensitivity of the human ear playing in tune is the paramount requirement when playing music. This implies, that embouchure actions provide an important insight into a subjective quality assessment of reed wind instruments from the viewpoint of the musician: An instrument requiring only small corrections will be comfortable to play. Theoretical investigations of the whole system of resonator, reed, and musician by use of a physical model nowadays still seem insufficient with respect to the required precision. Therefore the path of well-described artificial mouth measurements has been chosen here. For the separate treatment of the resonator and the double-reed, existing classical models have been used. Modifications to these models are suggested and verified experimentally. The influence of the musician is incorporated by the lip force-dependent initial reed slit height. For this investigation a measurement setup has been built that allows precise adjustment of lip force during playing. With measurements of the artificial mouth parameters blowing pressure, mouthpiece pressure, volume-flow rate and axial lip position on reed, the experiment is fully described for a given resonator setting represented by an input impedance curve. By use of the suggested empirical model the adjustment parameters can be turned into model parameters. A large data set from blowing experiments covering the full tonal and dynamical range on five modern German bassoons of different make is given and interpreted. The experimental data presented with this work can be a basis for extending the knowledge and understanding of the interaction of instrument, mouthpiece and player. On the one hand, they provide an objective insight into tuning aspects of the studied bassoons. On the other hand the experiments define working points of the coupled system by means of quasi-static model parameters. These may be useful to validate dynamical physical models in further studies. The experimental data provide an important prerequisite for scientific proposals of optimizations of the bassoon and other reed wind instruments. It can further serve as a fundament for the interdisciplinary communication between musicians, musical instrument makers and scientists.:1 Introduction 1 1.1 Motivation 1 1.2 Scientific Approaches to Woodwind Musical Instruments 3 1.3 Organization of the Thesis 6 2 Acoustical Properties of the Bassoon Air Column 7 2.1 Wave propagation in tubes 7 2.1.1 Theory 7 2.1.2 Transmission Line Modeling 8 2.1.3 Implementation 18 2.1.4 Remarks on Modeling Wall Losses in a Conical Waveguide 19 2.2 Input Impedance Measurement 23 2.2.1 Principle 23 2.2.2 Device 23 2.2.3 Calibration and Correction 24 2.3 Comparison of Theory and Experiment 27 2.3.1 Repeatability and Measurement Uncertainty 27 2.3.2 Comparison of numerical and experimental Impedance Curves 32 2.4 Harmonicity Analysis of the Resonator 35 2.4.1 The Role of the Resonator 35 2.4.2 The reed equivalent Volume 35 2.4.3 Harmonicity Map 36 2.5 Summary 38 3 Characterization of the Double Reed Mouthpiece 41 3.1 Physical Model of the Double-Reed 41 3.1.1 Working Principle 41 3.1.2 Structural Mechanical Characteristics 42 3.1.3 Fluid Mechanical Characteristics 44 3.2 Measurement of Reed Parameters 49 3.2.1 Quasi-stationary Measurement 49 3.2.2 Dynamic Measurement 50 3.3 Construction of an Artificial Mouth 52 3.3.1 Requirements Profile 52 3.3.2 Generic Design 53 3.3.3 The artificial Lip 54 3.3.4 Air Supply 55 3.3.5 Sensors and Data Acquisition 57 3.3.6 Experimental setup 59 3.4 Summary 59 4 Modeling Realistic Embouchures with Reed Parameters 61 4.1 Reed Channel Geometry and Flow Characteristics 61 4.1.1 The Double-Reed as a Flow Duct 61 4.1.2 Bernoulli Flow-Model with Pressure Losses 65 4.1.3 Discussion of the Model 68 4.2 Quasi-static Interaction of Flow and Reed-Channel 72 4.2.1 Pressure-driven Deformation of the Duct Intake 72 4.2.2 Reed-Flow Model including Channel Deformation 75 4.2.3 Influence of Model Parameters 76 4.2.4 Experimental Verification 78 4.3 Effect of the Embouchure on the Reed-Flow 81 4.3.1 Adjustment of the Initial Slit Height 81 4.3.2 Quasi-static Flow in the Deformed Reed-Channel 83 4.3.3 Simplified empirical Model including a Lip Force 85 4.4 Summary 93 5 Survey of Performance Characteristics of the Modern German Bassoon 5.1 Experimental Procedure and Data Analysis 95 5.1.1 Description of the Experiment 95 5.1.2 Time Domain Analysis 97 5.1.3 Spectral Analysis – Period Synchronized Sampling 98 5.1.4 Spectral Centroid and Formants 99 5.1.5 Embouchure parameters 100 5.2 Observations on the Bassoon under Operating Conditions 105 5.2.1 Excitation Parameter Ranges 106 5.2.2 Characteristics of the radiated Sound 110 5.2.3 Reed Pressure Waveform Analysis 115 5.2.4 Summarizing Overview 118 5.3 Performance Control with the Embouchure 120 5.3.1 Register-dependent Embouchure Characteristics 120 5.3.2 Intonation Corrections 123 5.3.3 Sound Color Adjustments 127 5.3.4 Relation to the acoustical Properties of the Resonator 129 5.4 Summary 137 6 Conclusion 139 6.1 Summary 139 6.2 Outlook 141