Paper Archive

The ubiquity of sound synthesizers have reshaped music production and even entirely define new music genres. However, the increasing complexity and number of parameters in modern synthesizers make them harder to master. Hence, the development of methods allowing to easily create and explore with synthesizers is a crucial need. Here, we introduce a radically novel formulation of audio synthesizer control by formalizing it as finding an organized continuous latent space of audio that represents the capabilities of a synthesizer and map this space to the space of synthesis parameter. By using this formulation, we show that we can address simultaneously automatic parameter inference, macro-control learning and audio-based preset exploration within a single model. To solve this new formulation, we rely on Variational Auto-Encoders (VAE) and Normalizing Flows (NF) to organize and map the respective auditory and parameter spaces. We introduce a new type of NF named regression flows that allow to perform an invertible mapping between separate latent spaces, while steering the organization of some of the latent dimensions. We evaluate our proposal against a large set of baseline models and show its superiority in both parameter inference and audio reconstruction. We also show that the model disentangles the major factors of audio variations as latent dimensions, that can be directly used as macro-parameters. Finally, we discuss the use of our model in several creative applications and introduce real-time implementations in Ableton Live

Download

Rigid-body impact sound synthesis methods often omit the ground sound. In this paper we analyze an idealized ground-sound model based on an elastodynamic halfspace, and use it to identify scenarios wherein ground sound is perceptually relevant versus when it is masked by the impacting object’s modal sound or transient acceleration noise. Our analytical model gives a smooth, closed-form expression for ground surface acceleration, which we can then use in the Rayleigh integral or in an “acoustic shader” for a finite-difference time-domain wave simulation. We find that when modal sound is inaudible, ground sound is audible in scenarios where a dense object impacts a soft ground and scenarios where the impact point has a low elevation angle to the listening point.

Download

Deep generative neural networks have thrived in the field of computer vision, enabling unprecedented intelligent image processes. Yet the results in audio remain less advanced and many applications are still to be investigated. Our project targets real-time sound synthesis from a reduced set of high-level parameters, including semantic controls that can be adapted to different sound libraries and specific tags. These generative variables should allow expressive modulations of target musical qualities and continuously mix into new styles. To this extent we train auto-encoders on an orchestral database of individual note samples, along with their intrinsic attributes: note class, timbre domain (an instrument subset) and extended playing techniques. We condition the decoder for explicit control over the rendered note attributes and use latent adversarial training for learning expressive style parameters that can ultimately be mixed. We evaluate both generative performances and correlations of the attributes with the latent representation. Our ablation study demonstrates the effectiveness of the musical conditioning. The proposed model generates individual notes as magnitude spectrograms from any probabilistic latent code samples (each latent point maps to a single note), with expressive control of orchestral timbres and playing styles. Its training data subsets can directly be visualized in the 3-dimensional latent representation. Waveform rendering can be done offline with the Griffin-Lim algorithm. In order to allow real-time interactions, we fine-tune the decoder with a pretrained magnitude spectrogram inversion network and embed the full waveform generation pipeline in a plugin. Moreover the encoder could be used to process new input samples, after manipulating their latent attribute representation, the decoder can generate sample variations as an audio effect would. Our solution remains rather light-weight and fast to train, it can directly be applied to other sound domains, including an user’s libraries with custom sound tags that could be mapped to specific generative controls. As a result, it fosters creativity and intuitive audio style experimentations. Sound examples and additional visualizations are available on Github1, as well as codes after the review process.

Download

This paper proposes a new method to interpolate between two audio signals. As an interpolation parameter is changed, the pitches in one signal slide to the pitches in the other, producing a portamento, or musical glide. The assignment of pitches in one sound to pitches in the other is accomplished by solving a 1-dimensional optimal transport problem. In addition, we introduce several techniques that preserve the audio fidelity over this highly nonlinear transformation. A portamento is a natural way for a musician to transition between notes, but traditionally it has only been possible for instruments with a continuously variable pitch like the human voice or the violin. Audio transport extends the portamento to any instrument, even polyphonic ones. Moreover, the effect can be used to transition between different instruments, groups of instruments, or any other pair of audio signals. The audio transport effect operates in real-time; we provide an open-source implementation. In experiments with sinusoidal inputs, the interpolating effect is indistinguishable from ideal sine sweeps. More generally, the effect produces clear, musical results for a wide variety of inputs.

Download

Nonlinear systems, like e.g. guitar distortion effects, play an important role in musical signal processing. One major problem encountered in digital nonlinear systems is aliasing distortion. Consequently, various aliasing reduction methods have been proposed in the literature. One of these is based on using the antiderivative of the nonlinearity and has proven effective, but is limited to memoryless systems. In this work, it is extended to a class of stateful systems which includes but is not limited to systems with a single one-port nonlinearity. Two examples from the realm of virtual analog modeling show its applicability to and effectiveness for commonly encountered guitar distortion effect circuits.

Download

For decades, analog magnetic tape recording was the most popular method for recording music, but has been replaced over the past 30 years first by DAT tape, then by DAWs and audio interfaces. Despite being replaced by higher quality technology, many have sought to recreate a "tape" sound through digital effects, despite the distortion, tape "hiss", and other oddities analog tape produced. The following paper describes the general process of creating a physical model of an analog tape machine starting from basic physical principles, then discusses in-depth a real-time implementation of a physical model of a Sony TC-260 tape machine."Whatever you now find weird, ugly, uncomfortable, and nasty about a new medium will surely become its signature. CD distortion, the jitteriness of digital video, the crap sound of 8-bit - all of these will be cherished and emulated as soon as they can be avoided." -Brian Eno.

Download

The following article introduces a new parametric synthesis algorithm for sound textures inspired by existing methods used for visual textures. Using a 2D Convolutional Neural Network (CNN), a sound signal is modified until the temporal cross-correlations of the feature maps of its log-spectrogram resemble those of a target texture. We show that the resulting synthesized sound signal is both different from the original and of high quality, while being able to reproduce singular events appearing in the original. This process is performed in the time domain, discarding the harmful phase recovery step which usually concludes synthesis performed in the time-frequency domain. It is also straightforward and flexible, as it does not require any fine tuning between several losses when synthesizing diverse sound textures. Synthesized spectrograms and sound signals are showcased, and a way of extending the synthesis in order to produce a sound of any length is also presented. We also discuss the choice of CNN, border effects in our synthesized signals and possible ways of modifying the algorithm in order to improve its current long computation time.

Download

We apply modal synthesis to create a virtual collection of crash cymbals. Synthesizing each cymbal may require enough modes to stress a modern CPU, so a full drum set would certainly not be tractable in real-time. To work around this, we create a GPU-accelerated modal filterbank, with each individual set piece allocated over two thousand modes. This takes only a fraction of available GPU floating-point throughput. With CPU resources freed up, we explore methods to model the different instrument response in the linear/harmonic and non-linear/inharmonic regions that occur as more energy is present in a cymbal: a simple approach, yet one that preserves the parallelism of the problem, uses multisampling, and a more physically-based approach approximates modal coupling.

Download