Download Differentiable Feedback Delay Network for Colorless Reverberation Artificial reverberation algorithms often suffer from spectral coloration, usually in the form of metallic ringing, which impairs the perceived quality of sound. This paper proposes a method to reduce the coloration in the feedback delay network (FDN), a popular artificial reverberation algorithm. An optimization framework is employed entailing a differentiable FDN to learn a set of parameters decreasing coloration. The optimization objective is to minimize the spectral loss to obtain a flat magnitude response, with an additional temporal loss term to control the sparseness of the impulse response. The objective evaluation of the method shows a favorable narrower distribution of modal excitation while retaining the impulse response density. The subjective evaluation demonstrates that the proposed method lowers perceptual coloration of late reverberation, and also shows that the suggested optimization improves sound quality for small FDN sizes. The method proposed in this work constitutes an improvement in the design of accurate and high-quality artificial reverberation, simultaneously offering computational savings.
Download Optimization techniques for a physical model of human vocalisation We present a non-supervised approach to optimize and evaluate the synthesis of non-speech audio effects from a speech production model. We use the Pink Trombone synthesizer as a case study of a simplified production model of the vocal tract to target nonspeech human audio signals –yawnings. We selected and optimized the control parameters of the synthesizer to minimize the difference between real and generated audio. We validated the most common optimization techniques reported in the literature and a specifically designed neural network. We evaluated several popular quality metrics as error functions. These include both objective quality metrics and subjective-equivalent metrics. We compared the results in terms of total error and computational demand. Results show that genetic and swarm optimizers outperform least squares algorithms at the cost of executing slower and that specific combinations of optimizers and audio representations offer significantly different results. The proposed methodology could be used in benchmarking other physical models and audio types.
Download Modulation Extraction for LFO-driven Audio Effects Low frequency oscillator (LFO) driven audio effects such as phaser, flanger, and chorus, modify an input signal using time-varying filters and delays, resulting in characteristic sweeping or widening effects. It has been shown that these effects can be modeled using neural networks when conditioned with the ground truth LFO signal. However, in most cases, the LFO signal is not accessible and measurement from the audio signal is nontrivial, hindering the modeling process. To address this, we propose a framework capable of extracting arbitrary LFO signals from processed audio across multiple digital audio effects, parameter settings, and instrument configurations. Since our system imposes no restrictions on the LFO signal shape, we demonstrate its ability to extract quasiperiodic, combined, and distorted modulation signals that are relevant to effect modeling. Furthermore, we show how coupling the extraction model with a simple processing network enables training of end-to-end black-box models of unseen analog or digital LFO-driven audio effects using only dry and wet audio pairs, overcoming the need to access the audio effect or internal LFO signal. We make our code available and provide the trained audio effect models in a real-time VST plugin1 .
Download A Common-Slopes Late Reverberation Model Based on Acoustic Radiance Transfer In rooms with complex geometry and uneven distribution of energy losses, late reverberation depends on the positions of sound sources and listeners. More precisely, the decay of energy is characterised by a sum of exponential curves with position-dependent amplitudes and position-independent decay rates (hence the name common slopes). The amplitude of different energy decay components is a particularly important perceptual aspect that requires efficient modeling in applications such as virtual reality and video games. Acoustic Radiance Transfer (ART) is a room acoustics model focused on late reverberation, which uses a pre-computed acoustic transfer matrix based on the room geometry and materials, and allows interactive changes to source and listener positions. In this work, we present an efficient common-slopes approximation of the ART model. Our technique extracts common slopes from ART using modal decomposition, retaining only the non-oscillating energy modes. Leveraging the structure of ART, changes to the positions of sound sources and listeners only require minimal processing. Experimental results show that even very few slopes are sufficient to capture the positional dependency of late reverberation, reducing model complexity substantially.
Download Interpolation Filters for Antiderivative Antialiasing Aliasing is an inherent problem in nonlinear digital audio processing which results in undesirable audible artefacts. Antiderivative antialiasing has proved to be an effective approach to mitigate aliasing distortion, and is based on continuous-time convolution of a linearly interpolated distorted signal with antialiasing filter kernels. However, the performance of this method is determined by the properties of interpolation filter. In this work, cubic interpolation kernels for antiderivative antialiasing are considered. For memoryless nonlinearities, aliasing reduction is improved employing cubic interpolation. For stateful systems, numerical simulation and stability analysis with respect to different interpolation kernels remain in favour of linear interpolation.
Download Real-Time Implementation of a Linear-Phase Octave Graphic Equalizer This paper proposes a real-time implementation of a linear-phase octave graphic equalizer (GEQ), previously introduced by the same authors. The structure of the GEQ is based on interpolated finite impulse response (IFIR) filters and is derived from a single prototype FIR filter. The low computational cost and small latency make the presented GEQ suitable for real-time applications. In this work, the GEQ has been implemented as a plugin of a specific software, used for real-time tests. The performance of the equalizer has been evaluated through subjective tests, comparing it with a filterbank equalizer. For the tests, four standard equalization curves have been chosen. The experimental results show promising outcomes. The result is an accurate real-time-capable linear-phase GEQ with a reasonable latency.
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 Inference-Time Structured Pruning for Real-Time Neural Network Audio Effects Structured pruning is a technique for reducing the computational
load and memory footprint of neural networks by removing structured subsets of parameters according to a predefined schedule
or ranking criterion.
This paper investigates the application of
structured pruning to real-time neural network audio effects, focusing on both feedforward networks and recurrent architectures.
We evaluate multiple pruning strategies at inference time, without retraining, and analyze their effects on model performance. To
quantify the trade-off between parameter count and audio fidelity,
we construct a theoretical model of the approximation error as a
function of network architecture and pruning level. The resulting bounds establish a principled relationship between pruninginduced sparsity and functional error, enabling informed deployment of neural audio effects in constrained real-time environments.
Download Unsupervised Text-to-Sound Mapping via Embedding Space Alignment This work focuses on developing an artistic tool that performs an
unsupervised mapping between text and sound, converting an input text string into a series of sounds from a given sound corpus.
With the use of a pre-trained sound embedding model and a separate, pre-trained text embedding model, the goal is to find a mapping between the two feature spaces. Our approach is unsupervised which allows any sound corpus to be used with the system.
The tool performs the task of text-to-sound retrieval, creating a
soundfile in which each word in the text input is mapped to a single sound in the corpus, and the resulting sounds are concatenated
to play sequentially. We experiment with three different mapping
methods, and perform quantitative and qualitative evaluations on
the outputs. Our results demonstrate the potential of unsupervised
methods for creative applications in text-to-sound mapping.
Download Low-cost Numerical Approximation of HRTFs: a Non-Linear Frequency Sampling Approach Head-related transfer functions (HRTFs) describe filters that model the scattering effect of the human body on sound waves. In their discrete-time form, they are used in acoustic simulations for virtual reality (VR) or augmented reality (AR), and since HRTFs are listener-specific, the use of individualized HRTFs allows achieving more realistic perceptual results. One way to produce individualized HRTFs is by estimating the sound field around the subjects’ 3D representations (meshes) via numerical simulations, which compute discrete complex pressure values in the frequency domain in regular frequency steps. Despite the advances in the area, the computational resources required for this process are still considerably high and increase with frequency. The goal of this paper is to tackle the high computational cost associated with this task by sampling the frequency domain using hybrid linear-logarithmic frequency resolution. The results attained in simulations performed using 23 real 3D meshes suggest that the proposed strategy is able to reduce the computational cost while still providing remarkably low spectral distortion, even in simulations that require as little as 11.2% of the original total processing time.