Research objectives and content
Our research project copes with the physical modelling of the piano instrument through the eyes of system identification: Given a consistent parametric model structure of the piano dynamics, relevant parameter values are estimated by minimising the deviation between true piano acoustic recordings and the corresponding model response.
Although the modal analysis of the vibrations of a single piano string is elementary, the modelling of the piano sound is intricate because of complex acoustic-mechanical coupling effects (see e.g. N. Fletcher et al., The physics of musical instruments, Springer-Verlag, New-York, 1991). Here, the string dynamics is a distributed system with two inputs and one output. The inputs are the hammer excitation and the bridge force at one of the string ends. The system output is the movement of the string at the bridge side. String vibrations occur when feeding back the system output at the second system input. A large effort will be devoted to carefully detail this feedback dynamics, i.e. coupling between the (in unison) strings and the bridge reaction. The hammer dynamics has a fundamental impact on the time-frequency properties of the piano tone so that a detailed mechanical modelling of the hammer action is needed. As a result, the physical model of the hammer-soundboard dynamics consists in a parametrised model structure whose l parameter are to be estimated on the basis of elementary piano tone records.
Training content (objective, benefit and expected impact)
The objectives and the benefits of the training consist in the physical modelling of mechanical systems, in the acoustic and | identification expertises as well as in the use of acoustically-oriented computer materials. The training impacts rely on the | elaboration of simulation models of the piano instrument and on the development of consequent sound synthetizers
Links with industry / industrial relevance (22)