Download Antiderivative Antialiasing for Recurrent Neural Networks Neural networks have become invaluable for general audio processing tasks, such as virtual analog modeling of nonlinear audio equipment.
For sequence modeling tasks in particular, recurrent neural networks (RNNs) have gained widespread adoption in recent years. Their general applicability and effectiveness
stems partly from their inherent nonlinearity, which makes them
prone to aliasing. Recent work has explored mitigating aliasing
by oversampling the network—an approach whose effectiveness is
directly linked with the incurred computational costs. This work
explores an alternative route by extending the antiderivative antialiasing technique to explicit, computable RNNs. Detailed applications to the Gated Recurrent Unit and Long Short-Term Memory cell are shown as case studies. The proposed technique is evaluated
on multiple pre-trained guitar amplifier models, assessing its impact on the amount of aliasing and model tonality. The method is
shown to reduce the models’ tendency to alias considerably across
all considered sample rates while only affecting their tonality moderately, without requiring high oversampling factors. The results
of this study can be used to improve sound quality in neural audio
processing tasks that employ a suitable class of RNNs. Additional
materials are provided in the accompanying webpage.
Download Principal Component Analysis of Rasterised Audio for Cross-Synthesis This paper describes a system for cross synthesis of rasterised time-domain audio. Rasterisation of the audio allows alignment of the macroscopic features of audio samples of instrument tones prior to principal component analysis (PCA). Specifically a novel algorithm for straightening and aligning rastogram features has been developed which is based on an interactive process incorporating the Canny detection algorithm and variable resampling. Timbral cross-synthesis is achieved by projecting a given instrument tone onto the principal components derived from a training set of sounds for a different tone. The alignment algorithm improves the efficiency of PCA for resynthesizing tones.
Download Modal Based Tanpura Simulation: Combining Tension Modulation and Distributed Bridge Interaction Techniques for the simulation of the tanpura have advanced significantly in recent years allowing numerically stable inclusion of bridge contact. In this paper tension modulation is added to a tanpura model containing a stiff lossy string, distributed bridge contact and the thread. The model is proven to be unconditionally stable and the numerical solver used has a unique solution as a result of choices made in the discretisation process. Effects due to the distribution of the bridge contact forces by comparison to a single point bridge and of introducing the tension modulation are studied in simulations. This model is intended for use in furthering the understanding of the physics of the tanpura and for informing the development of algorithms for sound synthesis of the tanpura and similar stringed instruments.
Download Solid State Bus-Comp: A Large-Scale and Diverse Dataset for Dynamic Range Compressor Virtual Analog Modeling Virtual Analog (VA) modeling aims to simulate the behavior
of hardware circuits via algorithms to replicate their tone digitally.
Dynamic Range Compressor (DRC) is an audio processing module
that controls the dynamics of a track by reducing and amplifying
the volumes of loud and quiet sounds, which is essential in music
production. In recent years, neural-network-based VA modeling has
shown great potential in producing high-fidelity models. However,
due to the lack of data quantity and diversity, their generalization
ability in different parameter settings and input sounds is still limited. To tackle this problem, we present Solid State Bus-Comp, the
first large-scale and diverse dataset for modeling the classical VCA
compressor — SSL 500 G-Bus. Specifically, we manually collected
175 unmastered songs from the Cambridge Multitrack Library. We
recorded the compressed audio in 220 parameter combinations,
resulting in an extensive 2528-hour dataset with diverse genres, instruments, tempos, and keys. Moreover, to facilitate the use of our
proposed dataset, we conducted benchmark experiments in various
open-sourced black-box and grey-box models, as well as white-box
plugins. We also conducted ablation studies in different data subsets to illustrate the effectiveness of the improved data diversity and
quantity. The dataset and demos are on our project page: https:
//www.yichenggu.com/SolidStateBusComp/.
Download Fourier Neural Operators for Sample-Rate-Independent Virtual Analog Modeling Neural networks that operate directly on time-domain signals are widely used for virtual analog (VA) modeling. A key limitation of these models is their dependence on the sampling rate used during training, which becomes implicitly encoded in the learned parameters, so that changing it generally alters the realized dynamics. Although architectural modifications to recurrent neural networks have been proposed to enable sample-rate independent operation, these approaches are inherently tailored to upsampling and do not accommodate downsampling scenarios. In this manuscript, we present a VA modeling framework based on Fourier Neural Operators (FNOs) adapted to process fixed-duration audio frames. The proposed formulation defines the learned mapping over a fixed temporal support and evaluates it on uniform grids of different densities, so that a model trained at a single sampling rate can be applied at unseen sampling resolutions. Numerical results on a nonlinear transistor circuit show that the proposed model achieves competitive accuracy in upsampling scenarios while remaining directly applicable to downsampling, unlike a sample-rate independent baseline recurrent architecture.
Download WildFX: A DAW-Powered Pipeline for In-the-Wild Audio FX Graph Modeling We present WildFX, a digital-audio-workstation-powered pipeline for modeling audio-effects graphs from in-the-wild audio. The system uses a DAW environment to construct, render, and evaluate effect-processing graphs, enabling research on realistic effect chains beyond isolated processors or synthetic training settings. WildFX supports the analysis and reconstruction of complex audio transformations by combining flexible plugin routing with data-driven modeling. The pipeline is designed to facilitate scalable dataset creation and experimentation with effect graph inference, parameter estimation, and audio transformation in practical production contexts.
Download Low-order allpass Interpolated Delay Loops This paper presents empirical and theoretical results for a delay line cascaded with a second-order allpass filter in a feedback loop. Though such a structure has been used for years to model stiff vibrating strings, the complete range of behavior of such a structure has not been fully described and analyzed. As shown in this paper, in addition to the desired behavior of providing a frequencydependent delay line length, other phenomena may occur, such as “beating” or “mode splitting.” Associated analysis simulation results are presented.
Download Phase-based informed source separation for active listening of music This paper presents an informed source separation technique of monophonic mixtures. Although the vast majority of the separation methods are based on the time-frequency energy of each source, we introduce a new approach using solely phase information to perform the separation. The sources are iteratively reconstructed using an adaptation of the Multiple Input Spectrogram Inversion (MISI) algorithm from Gunawan and Sen. The proposed method is then tested against conventional MISI and Wiener filtering on monophonic signals and oracle conditions. Results show that at the cost of a larger computation time, our method outperforms both MISI and Wiener filtering in oracle conditions with much higher objective quality even with phase quantization.
Download Modelling digital musical effects for signal processors, based on real effect manifestation analysis For quite some time in the area of commercial utilization of digital audio effects, efforts have emerged to create simulate by software analog effects and effect processors This paper deals with the analysis of musical effects, the design of algorithms for simulating these effects, and their realization on both digital signal processors and the PC platform in the form of plug-in modules for the DirectX environment. It also deals with the problem of controlling the effect parameters and with subjective testing of algorithms, and it examines the fidelity of simulated effects as compared with the original.
Download Audio-Based Gesture Extraction on the ESITAR Controller Using sensors to extract gestural information for control parameters of digital audio effects is common practice. There has also been research using machine learning techniques to classify specific gestures based on audio feature analysis. In this paper, we will describe our experiments in training a computer to map the appropriate audio-based features to look like sensor data, in order to potentially eliminate the need for sensors. Specifically, we will show our experiments using the ESitar, a digitally enhanced sensor based controller modeled after the traditional North Indian sitar. We utilize multivariate linear regression to map continuous audio features to continuous gestural data.