Download Modeling the Frequency-Dependent Sound Energy Decay of Acoustic Environments with Differentiable Feedback Delay Networks Differentiable machine learning techniques have recently proved effective for finding the parameters of Feedback Delay Networks (FDNs) so that their output matches desired perceptual qualities of target room impulse responses. However, we show that existing methods tend to fail at modeling the frequency-dependent behavior of sound energy decay that characterizes real-world environments unless properly trained. In this paper, we introduce a novel perceptual loss function based on the mel-scale energy decay relief, which generalizes the well-known time-domain energy decay curve to multiple frequency bands. We also augment the prototype FDN by incorporating differentiable wideband attenuation and output filters, and train them via backpropagation along with the other model parameters. The proposed approach improves upon existing strategies for designing and training differentiable FDNs, making it more suitable for audio processing applications where realistic and controllable artificial reverberation is desirable, such as gaming, music production, and virtual reality.
Download Sample Rate Independent Recurrent Neural Networks for Audio Effects Processing In recent years, machine learning approaches to modelling guitar amplifiers and effects pedals have been widely investigated and have become standard practice in some consumer products. In particular, recurrent neural networks (RNNs) are a popular choice for modelling non-linear devices such as vacuum tube amplifiers and distortion circuitry. One limitation of such models is that they are trained on audio at a specific sample rate and therefore give unreliable results when operating at another rate. Here, we investigate several methods of modifying RNN structures to make them approximately sample rate independent, with a focus on oversampling. In the case of integer oversampling, we demonstrate that a previously proposed delay-based approach provides high fidelity sample rate conversion whilst additionally reducing aliasing. For non-integer sample rate adjustment, we propose two novel methods and show that one of these, based on cubic Lagrange interpolation of a delay-line, provides a significant improvement over existing methods. To our knowledge, this work provides the first in-depth study into this problem.
Download DataRES and PyRES: A Room Dataset and a Python Library for Reverberation Enhancement System Development, Evaluation, and Simulation Reverberation is crucial in the acoustical design of physical
spaces, especially halls for live music performances. Reverberation Enhancement Systems (RESs) are active acoustic systems that
can control the reverberation properties of physical spaces, allowing them to adapt to specific acoustical needs. The performance of
RESs strongly depends on the properties of the physical room and
the architecture of the Digital Signal Processor (DSP). However,
room-impulse-response (RIR) measurements and the DSP code
from previous studies on RESs have never been made open access, leading to non-reproducible results. In this study, we present
DataRES and PyRES—a RIR dataset and a Python library to increase the reproducibility of studies on RESs. The dataset contains RIRs measured in RES research and development rooms and
professional music venues. The library offers classes and functionality for the development, evaluation, and simulation of RESs.
The implemented DSP architectures are made differentiable, allowing their components to be trained in a machine-learning-like
pipeline. The replication of previous studies by the authors shows
that PyRES can become a useful tool in future research on RESs.
Download Differentiable Active Acoustics - Optimizing Stability via Gradient Descent Active acoustics (AA) refers to an electroacoustic system that actively modifies the acoustics of a room. For common use cases, the number of transducers—loudspeakers and microphones—involved in the system is large, resulting in a large number of system parameters. To optimally blend the response of the system into the natural acoustics of the room, the parameters require careful tuning, which is a time-consuming process performed by an expert. In this paper, we present a differentiable AA framework, which allows multi-objective optimization without impairing architecture flexibility. The system is implemented in PyTorch to be easily translated into a machine-learning pipeline, thus automating the tuning process. The objective of the pipeline is to optimize the digital signal processor (DSP) component to evenly distribute the energy in the feedback loop across frequencies. We investigate the effectiveness of DSPs composed of finite impulse response filters, which are unconstrained during the optimization. We study the effect of multiple filter orders, number of transducers, and loss functions on the performance. Different loss functions behave similarly for systems with few transducers and low-order filters. Increasing the number of transducers and the order of the filters improves results and accentuates the difference in the performance of the loss functions.
Download Learning Nonlinear Dynamics in Physical Modelling Synthesis Using Neural Ordinary Differential Equations Modal synthesis methods are a long-standing approach for modelling distributed musical systems. In some cases extensions are
possible in order to handle geometric nonlinearities. One such
case is the high-amplitude vibration of a string, where geometric nonlinear effects lead to perceptually important effects including pitch glides and a dependence of brightness on striking amplitude. A modal decomposition leads to a coupled nonlinear system of ordinary differential equations. Recent work in applied machine learning approaches (in particular neural ordinary differential equations) has been used to model lumped dynamic systems
such as electronic circuits automatically from data. In this work,
we examine how modal decomposition can be combined with neural ordinary differential equations for modelling distributed musical systems. The proposed model leverages the analytical solution
for linear vibration of system’s modes and employs a neural network to account for nonlinear dynamic behaviour. Physical parameters of a system remain easily accessible after the training without
the need for a parameter encoder in the network architecture. As
an initial proof of concept, we generate synthetic data for a nonlinear transverse string and show that the model can be trained to
reproduce the nonlinear dynamics of the system. Sound examples
are presented.
Download Exposure Bias and State Matching in Recurrent Neural Network Virtual Analog Models Virtual analog (VA) modeling using neural networks (NNs) has
great potential for rapidly producing high-fidelity models. Recurrent neural networks (RNNs) are especially appealing for VA due
to their connection with discrete nodal analysis. Furthermore, VA
models based on NNs can be trained efficiently by directly exposing them to the circuit states in a gray-box fashion. However,
exposure to ground truth information during training can leave the
models susceptible to error accumulation in a free-running mode,
also known as “exposure bias” in machine learning literature. This
paper presents a unified framework for treating the previously
proposed state trajectory network (STN) and gated recurrent unit
(GRU) networks as special cases of discrete nodal analysis. We
propose a novel circuit state-matching mechanism for the GRU
and experimentally compare the previously mentioned networks
for their performance in state matching, during training, and in exposure bias, during inference. Experimental results from modeling
a diode clipper show that all the tested models exhibit some exposure bias, which can be mitigated by truncated backpropagation
through time. Furthermore, the proposed state matching mechanism improves the GRU modeling performance of an overdrive
pedal and a phaser pedal, especially in the presence of external
modulation, apparent in a phaser circuit.
Download Identification of individual guitar sounds by support vector machines This paper introduces an automatic classification system for the identification of individual classical guitars by single notes played on these guitars. The classification is performed by Support Vector Machines (SVM) that have been trained with the features of the single notes. The features used for classification were the time series of the partial tones, the time series of the MFCCs (Mel Frequency Cepstral Coefficients), and the “nontonal” contributions to the spectrum. The influences of these features on the classification success are reported. With this system, 80% of the sounds recorded with three different guitars were classified correctly. A supplementary classification experiment was carried out with human listeners resulting in a rate of 65% of correct classifications.
Download A Hierarchical Deep Learning Approach for Minority Instrument Detection Identifying instrument activities within audio excerpts is vital in music information retrieval, with significant implications for music cataloging and discovery. Prior deep learning endeavors in musical instrument recognition have predominantly emphasized instrument classes with ample data availability. Recent studies have demonstrated the applicability of hierarchical classification in detecting instrument activities in orchestral music, even with limited fine-grained annotations at the instrument level. Based on the Hornbostel-Sachs classification, such a hierarchical classification system is evaluated using the MedleyDB dataset, renowned for its diversity and richness concerning various instruments and music genres. This work presents various strategies to integrate hierarchical structures into models and tests a new class of models for hierarchical music prediction. This study showcases more reliable coarse-level instrument detection by bridging the gap between detailed instrument identification and group-level recognition, paving the way for further advancements in this domain.
Download Hubness-Aware Outlier Detection for Music Genre Recognition Outlier detection is the task of automatic identification of unknown data not covered by training data (e.g. a new genre in genre recognition). We explore outlier detection in the presence of hubs and anti-hubs, i.e. data objects which appear to be either very close or very far from most other data due to a problem of measuring distances in high dimensions. We compare a classic distance based method to two new approaches, which have been designed to counter the negative effects of hubness, on two standard music genre data sets. We demonstrate that anti-hubs are responsible for many detection errors and that this can be improved by using a hubness-aware approach.
Download Neural Net Tube Models for Wave Digital Filters Herein, we demonstrate the use of neural nets towards simulating multiport nonlinearities inside a wave digital filter. We introduce a resolved wave definition which allows us to extract features from a Kirchhoff domain dataset and train our neural networks directly in the wave domain. A hyperparameter search is performed to minimize error and runtime complexity. To illustrate the method, we model a tube amplifier circuit inspired by the preamplifier stage of the Fender Pro-Junior guitar amplifier. We analyze the performance of our neural nets models by comparing their distortion characteristics and transconductances. Our results suggest that activation function selection has a significant effect on the distortion characteristic created by the neural net.