Paper Archive

Neural audio synthesis methods can achieve high-fidelity and realistic sound generation by utilizing deep generative models. Such models typically rely on external labels which are often discrete as conditioning information to achieve guided sound generation. However, it remains difficult to control the subtle changes in sounds without appropriate and descriptive labels, especially given a limited dataset. This paper proposes an implicit conditioning method for neural audio synthesis using generative adversarial networks that allows for interpretable control of the acoustic features of synthesized sounds. Our technique creates a continuous conditioning space that enables timbre manipulation without relying on explicit labels. We further introduce an evaluation metric to explore controllability and demonstrate that our approach is effective in enabling a degree of controlled variation of different synthesized sound effects for in-domain and cross-domain sounds.

Download

This paper seeks to improve the state-of-the-art in delay-networkbased analysis-synthesis of measured room impulse responses (RIRs). We propose an informed method incorporating improved energy decay estimation and synthesis with an optimized feedback delay network. The performance of the presented method is compared against an end-to-end deep-learning approach. A formal listening test was conducted where participants assessed the similarity of reverberated material across seven distinct RIRs and three different sound sources. The results reveal that the performance of these methods is influenced by both the excitation sounds and the reverberation conditions. Nonetheless, the proposed method consistently demonstrates higher similarity ratings compared to the end-to-end approach across most conditions. However, achieving an indistinguishable synthesis of measured RIRs remains a persistent challenge, underscoring the complexity of this problem. Overall, this work helps improve the sound quality of analysis-based artificial reverberation.

Download

Deep learning models have seen widespread use in modelling LFOdriven audio effects, such as phaser and flanger. Although existing neural architectures exhibit high-quality emulation of individual effects, they do not possess the capability to manipulate the output via control parameters. To address this issue, we introduce Controllable Neural Frame-based Modulation Effects (CONMOD), a single black-box model which emulates various LFOdriven effects in a frame-wise manner, offering control over LFO frequency and feedback parameters. Additionally, the model is capable of learning the continuous embedding space of two distinct phaser effects, enabling us to steer between effects and achieve creative outputs. Our model outperforms previous work while possessing both controllability and universality, presenting opportunities to enhance creativity in modern LFO-driven audio effects. Additional demo of our model is available in the accompanying website.1

Download

Reverberation is a key element in spatial audio perception, historically achieved with the use of analogue devices, such as plate and spring reverb, and in the last decades with digital signal processing techniques that have allowed different approaches for Virtual Analogue Modelling (VAM). The electromechanical functioning of the spring reverb makes it a nonlinear system that is difficult to fully emulate in the digital domain with white-box modelling techniques. In this study, we compare five different neural network architectures, including convolutional and recurrent models, to assess their effectiveness in replicating the characteristics of this audio effect. The evaluation is conducted on two datasets at sampling rates of 16 kHz and 48 kHz. This paper specifically focuses on neural audio architectures that offer parametric control, aiming to advance the boundaries of current black-box modelling techniques in the domain of spring reverberation.

Download

Neural networks have found application within the Wave Digital Filters (WDFs) framework as data-driven input-output blocks for modeling single one-port or multi-port nonlinear devices in circuit systems. However, traditional neural networks lack predictable bounds for their output derivatives, essential to ensure convergence when simulating circuits with multiple nonlinear elements using fixed-point iterative methods, e.g., the Scattering Iterative Method (SIM). In this study, we address such issue by employing Lipschitz-bounded neural networks for regressing nonlinear WD scattering relations of one-port nonlinearities.

Download

A Head-Related Transfer Function (HRTF) is able to capture alterations a sound wave undergoes from its source before it reaches the entrances of a listener’s left and right ear canals, and is imperative for creating immersive experiences in virtual and augmented reality (VR/AR). Nevertheless, creating personalized HRTFs demands sophisticated equipment and is hindered by time-consuming data acquisition processes. To counteract these challenges, various techniques for HRTF interpolation and up-sampling have been proposed. This paper illustrates how Generative Adversarial Networks (GANs) can be applied to HRTF data upsampling in the spherical harmonics domain. We propose using Autoencoding Generative Adversarial Networks (AE-GAN) to upsample lowdegree spherical harmonics coefficients and get a more accurate representation of the full HRTF set. The proposed method is benchmarked against two baselines: barycentric interpolation and HRTF selection. Results from log-spectral distortion (LSD) evaluation suggest that the proposed AE-GAN has significant potential for upsampling very sparse HRTFs, achieving 17% improvement over baseline methods.

Download

Object-based audio production requires the positional metadata to be defined for each point-source object, including the key elements in the foreground of the sound scene. In many media production use cases, both cameras and microphones are employed to make recordings, and the human voice is often a key element. In this research, we detect and locate the active speaker in the video, facilitating the automatic extraction of the positional metadata of the talker relative to the camera’s reference frame. With the integration of the visual modality, this study expands upon our previous investigation focused solely on audio-based active speaker detection and localization. Our experiments compare conventional audio-visual approaches for active speaker detection that leverage monaural audio, our previous audio-only method that leverages multichannel recordings from a microphone array, and a novel audio-visual approach integrating vision and multichannel audio. We found the role of the two modalities to complement each other. Multichannel audio, overcoming the problem of visual occlusions, provides a double-digit reduction in detection error compared to audio-visual methods with single-channel audio. The combination of multichannel audio and vision further enhances spatial accuracy, leading to a four-percentage point increase in F1 score on the Tragic Talkers dataset. Future investigations will assess the robustness of the model in noisy and highly reverberant environments, as well as tackle the problem of off-screen speakers.

Download

Sound matching allows users to automatically approximate existing sounds using a synthesizer. Previous work has mostly focused on algorithms for automatically programming an existing synthesizer. This paper proposes a system for selecting between different synthesizer designs, each one with a corresponding automatic programmer. An implementation that allows designing ensembles based on a template is demonstrated. Several experiments are presented using a simple subtractive synthesis design. Using an ensemble of synthesizer-programmer pairs is shown to provide better matching than a single programmer trained for an equivalent integrated synthesizer. Scaling to hundreds of synthesizers is shown to improve match quality.

Download

Recent years have seen increasing interest in applying deep learning methods to the modeling of guitar amplifiers or effect pedals. Existing methods are mainly based on the supervised approach, requiring temporally-aligned data pairs of unprocessed and rendered audio. However, this approach does not scale well, due to the complicated process involved in creating the data pairs. A very recent work done by Wright et al. has explored the potential of leveraging unpaired data for training, using a generative adversarial network (GAN)-based framework. This paper extends their work by using more advanced discriminators in the GAN, and using more unpaired data for training. Specifically, drawing inspiration from recent advancements in neural vocoders, we employ in our GANbased model for guitar amplifier modeling two sets of discriminators, one based on multi-scale discriminator (MSD) and the other multi-period discriminator (MPD). Moreover, we experiment with adding unprocessed audio signals that do not have the corresponding rendered audio of a target tone to the training data, to see how much the GAN model benefits from the unpaired data. Our experiments show that the proposed two extensions contribute to the modeling of both low-gain and high-gain guitar amplifiers.

Download

In this paper we propose a method that can address a novel task, audio effect (AFX) chain estimation and dry signal recovery. AFXs are indispensable in modern sound design workflows. Sound engineers often cascade different AFXs (as an AFX chain) to achieve their desired soundscapes. Given a multi-AFX-applied solo instrument performance (wet signal), our method can automatically estimate the applied AFX chain and recover its unprocessed dry signal, while previous research only addresses one of them. The estimated chain is useful for novice engineers in learning practical usages of AFXs, and the recovered signal can be reused with a different AFX chain. To solve this task, we first develop a deep neural network model that estimates the last-applied AFX and undoes its AFX at a time. We then iteratively apply the same model to estimate the AFX chain and eventually recover the dry signal from the wet signal. Our experiments on guitar phrase recordings with various AFX chains demonstrate the validity of our method for both the AFX-chain estimation and dry signal recovery. We also confirm that the input wet signal can be reproduced by applying the estimated AFX chain to the recovered dry signal.

Download