Parametric Optimization of Violin Top Plates using Machine Learning

Abstract

We recently developed a neural network that receives as input the geometrical and mechanical parameters that define a violin top plate and gives as output its first ten eigenfrequencies computed in free boundary conditions. In this manuscript, we use the network to optimize several error functions, with the goal of analyzing the relationship between the eigenspectrum problem for violin top plates and their geometry. First, we focus on the violin outline. Given a vibratory feature, we find which is the best geometry of the plate to obtain it. Second, we investigate whether, from the vibrational point of view, a change in the outline shape can be compensated by one in the thickness distribution and vice versa. Finally, we analyze how to modify the violin shape to keep its response constant as its material properties vary. This is an original technique in musical acoustics, where artificial intelligence is not widely used yet. It allows us to both compute the vibrational behavior of an instrument from its geometry and optimize its shape for a given response. Furthermore, this method can be of great help to violin makers, who can thus easily understand the effects of the geometry changes in the violins they build, shedding light on one of the most relevant and, at the same time, less understood aspects of the construction process of musical instruments.

Predicting the Acoustics of Archtop Guitars using an AI-based Algorithm trained on FEM Simulations

“…Our FEM simulations are based on our previous studies [2,3,[10][11][12] with added complexity implied by the air volume inside and outside the body of the instrument, both defined through FEM and applying spherical wave radiation boundary on the external air domain. We already implemented the air in [12] but in that case the external air was modeled through Boundary Element Method (BEM).…”

“…In this context, using artificial intelligence (AI) to predict the acoustic behaviour of the digital model appears as a great opportunity to simplify the workflow [2][3][4]. Not only is it orders of magnitude faster than traditional FEM simulations, but once trained, the AI can be deployed in easy-to-use applications that do not require large computational power nor specialised knowledge from the user to be run.…”

Near field Acoustic Holography on arbitrary shapes using Convolutional Neural Network

“…The dataset is composed of 1568 different synthetic meshes of violin top plates with constant thickness and arching generated by the parametric model introduced in [29], [30]. The parameters defining the shapes are varied according to Gaussian distributions centered around the parameters of a reference violin, as described by the authors in [29], [30]. We computed the vibration and the radiated acoustic pressure of the plates through finite element analysis using COMSOL Multiphysics ® [31].…”

“…Lastly, we investigated the robustness of the proposed SRCNN against noisy input data. We tested the model with 6 test sets corrupted by different realizations of additive white gaussian noise with varying SNR ∈ [5,30] dB. Fig.…”