Download Introducing Deep Machine Learning for Parameter Estimation in Physical Modelling One of the most challenging tasks in physically-informed sound synthesis is the estimation of model parameters to produce a desired timbre. Automatic parameter estimation procedures have been developed in the past for some specific parameters or application scenarios but, up to now, no approach has been proved applicable to a wide variety of use cases. A general solution to parameters estimation problem is provided along this paper which is based on a supervised convolutional machine learning paradigm. The described approach can be classified as “end-to-end” and requires, thus, no specific knowledge of the model itself. Furthermore, parameters are learned from data generated by the model, requiring no effort in the preparation and labeling of the training dataset. To provide a qualitative and quantitative analysis of the performance, this method is applied to a patented digital waveguide pipe organ model, yielding very promising results.
Download Development of a Quality Assurance Automatic Listening Machine (QuAALM) This paper describes the development and application of a machine listening system for the automated testing of implementation equivalence in music signal processing effects which contain a high level of randomized time variation. We describe a mathematical model of generalized randomization in audio effects and explore different representations of the effect’s data. We then propose a set of classifiers to reliably determine if two implementations of the same randomized audio effect are functionally equivalent. After testing these classifiers against each other and against a set of human listeners we find the best implementation and determine that it agrees with the judgment of human listeners with an F1-Score of 0.8696.
Download Unsupervised Taxonomy of Sound Effects Sound effect libraries are commonly used by sound designers in a range of industries. Taxonomies exist for the classification of sounds into groups based on subjective similarity, sound source or common environmental context. However, these taxonomies are not standardised, and no taxonomy based purely on the sonic properties of audio exists. We present a method using feature selection, unsupervised learning and hierarchical clustering to develop an unsupervised taxonomy of sound effects based entirely on the sonic properties of the audio within a sound effect library. The unsupervised taxonomy is then related back to the perceived meaning of the relevant audio features.
Download Latent Force Models for Sound: Learning Modal Synthesis Parameters and Excitation Functions from Audio Recordings Latent force models are a Bayesian learning technique that combine physical knowledge with dimensionality reduction — sets of coupled differential equations are modelled via shared dependence on a low-dimensional latent space. Analogously, modal sound synthesis is a technique that links physical knowledge about the vibration of objects to acoustic phenomena that can be observed in data. We apply latent force modelling to sinusoidal models of audio recordings, simultaneously inferring modal synthesis parameters (stiffness and damping) and the excitation or contact force required to reproduce the behaviour of the observed vibrational modes. Exposing this latent excitation function to the user constitutes a controllable synthesis method that runs in real time and enables sound morphing through interpolation of learnt parameters.
Download Automatic Control of the Dynamic Range Compressor Using a Regression Model and a Reference Sound Practical experience with audio effects as well as knowledge of their parameters and how they change the sound is crucial when controlling digital audio effects. This often presents barriers for musicians and casual users in the application of effects. These users are more accustomed to describing the desired sound verbally or using examples, rather than understanding and configuring low-level signal processing parameters. This paper addresses this issue by providing a novel control method for audio effects. While a significant body of works focus on the use of semantic descriptors and visual interfaces, little attention has been given to an important modality, the use of sound examples to control effects. We use a set of acoustic features to capture important characteristics of sound examples and evaluate different regression models that map these features to effect control parameters. Focusing on dynamic range compression, results show that our approach provides a promising first step in this direction.
Download The Mix Evaluation Dataset Research on perception of music production practices is mainly concerned with the emulation of sound engineering tasks through lab-based experiments and custom software, sometimes with unskilled subjects. This can improve the level of control, but the validity, transferability, and relevance of the results may suffer from this artificial context. This paper presents a dataset consisting of mixes gathered in a real-life, ecologically valid setting, and perceptual evaluation thereof, which can be used to expand knowledge on the mixing process. With 180 mixes including parameter settings, close to 5000 preference ratings and free-form descriptions, and a diverse range of contributors from five different countries, the data offers many opportunities for music production analysis, some of which are explored here. In particular, more experienced subjects were found to be more negative and more specific in their assessments of mixes, and to increasingly agree with each other.
Download A Comparison of Player Performance in a Gamified Localisation Task Between Spatial Loudspeaker Systems This paper presents an experiment comparing player performance in a gamified localisation task between three loudspeaker configurations: stereo, 7.1 surround-sound and an equidistantly spaced octagonal array. The test was designed as a step towards determining whether spatialised game audio can improve player performance in a video game, thus influencing their overall experience. The game required players to find as many sound sources as possible, by using only sonic cues, in a 3D virtual game environment. Results suggest that the task was significantly easier when listening over a 7.1 surround-sound system, based on feedback from 24 participants. 7.1 was also the most preferred of the three listening conditions. The result was not entirely expected in that the octagonal array did not outperform 7.1. It is thought that, for the given stimuli, this may be a repercussion due to the octagonal array sacrificing an optimal front stereo pair, for more consistent imaging all around the listening space.
Download A Nonlinear Method for Manipulating Warmth and Brightness In musical timbre, two of the most commonly used perceptual dimensions are warmth and brightness. In this study, we develop a model capable of accurately controlling the warmth and brightness of an audio source using a single parameter. To do this, we first identify the most salient audio features associated with the chosen descriptors by applying dimensionality reduction to a dataset of annotated timbral transformations. Here, strong positive correlations are found between the centroid of various spectral representations and the most salient principal components. From this, we build a system designed to manipulate the audio features directly using a combination of linear and nonlinear processing modules. To validate the model, we conduct a series of subjective listening tests, and show that up to 80% of participants are able to allocate the correct term, or synonyms thereof, to a set of processed audio samples. Objectively, we show low Mahalanobis distances between the processed samples and clusters of the same timbral adjective in the low-dimensional timbre space.
Download Nicht-negativeMatrixFaktorisierungnutzendes-KlangsynthesenSystem (NiMFKS): Extensions of NMF-based Concatenative Sound Synthesis Concatenative sound synthesis (CSS) entails synthesising a “target” sound with other sounds collected in a “corpus.” Recent work explores CSS using non-negative matrix factorisation (NMF) to approximate a target sonogram by the product of a corpus sonogram and an activation matrix. In this paper, we propose a number of extensions of NMF-based CSS and present an open MATLAB implementation in a GUI-based application we name NiMFKS. In particular we consider the following extensions: 1) we extend the NMF framework by implementing update rules based on the generalised β-divergence; 2) We add an optional monotonic algorithm for sparse-NMF; 3) we tackle the computational challenges of scaling to big corpora by implementing a corpus pruning preprocessing step; 4) we generalise constraints that may be applied to the activation matrix shape; and 5) we implement new modes of interacting with the procedure by enabling sketching and modifying of the activation matrix. Our application, NiMFKS and source code can be downloaded from here: https: //code.soundsoftware.ac.uk/projects/nimfks.
Download Harmonic-percussive Sound Separation Using Rhythmic Information from Non-negative Matrix Factorization in Single-channel Music Recordings This paper proposes a novel method for separating harmonic and percussive sounds in single-channel music recordings. Standard non-negative matrix factorization (NMF) is used to obtain the activations of the most representative patterns active in the mixture. The basic idea is to classify automatically those activations that exhibit rhythmic and non-rhythmic patterns. We assume that percussive sounds are modeled by those activations that exhibit a rhythmic pattern. However, harmonic and vocal sounds are modeled by those activations that exhibit a less rhythmic pattern. The classification of the harmonic or percussive NMF activations is performed using a recursive process based on successive correlations applied to the activations. Specifically, promising results are obtained when a sound is classified as percussive through the identification of a set of peaks in the output of the fourth correlation. The reason is because harmonic sounds tend to be represented by one valley in a half-cycle waveform at the output of the fourth correlation. Evaluation shows that the proposed method provides competitive results compared to other reference state-of-the-art methods. Some audio examples are available to illustrate the separation performance of the proposed method.