Download Adversarial Synthesis of Drum Sounds Recent advancements in generative audio synthesis have allowed for the development of creative tools for generation and
manipulation of audio. In this paper, a strategy is proposed for the
synthesis of drum sounds using generative adversarial networks
(GANs). The system is based on a conditional Wasserstein GAN,
which learns the underlying probability distribution of a dataset
compiled of labeled drum sounds. Labels are used to condition
the system on an integer value that can be used to generate audio
with the desired characteristics. Synthesis is controlled by an input
latent vector that enables continuous exploration and interpolation
of generated waveforms. Additionally we experiment with a training method that progressively learns to generate audio at different
temporal resolutions. We present our results and discuss the benefits of generating audio with GANs along with sound examples
and demonstrations.
Download Self-Supervised Disentanglement of Harmonic and Rhythmic Features in Music Audio Signals The aim of latent variable disentanglement is to infer the multiple informative latent representations that lie behind a data generation process and is a key factor in controllable data generation. In this paper, we propose a deep neural network-based self-supervised learning method to infer the disentangled rhythmic and harmonic representations behind music audio generation. We train a variational autoencoder that generates an audio mel-spectrogram from two latent features representing the rhythmic and harmonic content. In the training phase, the variational autoencoder is trained to reconstruct the input mel-spectrogram given its pitch-shifted version. At each forward computation in the training phase, a vector rotation operation is applied to one of the latent features, assuming that the dimensions of the feature vectors are related to pitch intervals. Therefore, in the trained variational autoencoder, the rotated latent feature represents the pitch-related information of the mel-spectrogram, and the unrotated latent feature represents the pitch-invariant information, i.e., the rhythmic content. The proposed method was evaluated using a predictor-based disentanglement metric on the learned features. Furthermore, we demonstrate its application to the automatic generation of music remixes.
Download Anti-Aliasing of Neural Distortion Effects via Model Fine Tuning Neural networks have become ubiquitous with guitar distortion
effects modelling in recent years. Despite their ability to yield
perceptually convincing models, they are susceptible to frequency
aliasing when driven by high frequency and high gain inputs.
Nonlinear activation functions create both the desired harmonic
distortion and unwanted aliasing distortion as the bandwidth of
the signal is expanded beyond the Nyquist frequency. Here, we
present a method for reducing aliasing in neural models via a
teacher-student fine tuning approach, where the teacher is a pretrained model with its weights frozen, and the student is a copy of
this with learnable parameters. The student is fine-tuned against
an aliasing-free dataset generated by passing sinusoids through
the original model and removing non-harmonic components from
the output spectra.
Our results show that this method significantly suppresses aliasing for both long-short-term-memory networks (LSTM) and temporal convolutional networks (TCN). In the
majority of our case studies, the reduction in aliasing was greater
than that achieved by two times oversampling. One side-effect
of the proposed method is that harmonic distortion components
are also affected.
This adverse effect was found to be modeldependent, with the LSTM models giving the best balance between
anti-aliasing and preserving the perceived similarity to an analog
reference device.
Download Interpretation and control in AM/FM-based audio effects This paper is a continuation of our first studies on AM/FM digital audio effects, where the AM/FM decomposition equations were reviewed and some exploratory examples of effects were introduced. In the current paper we present more insight on the signals obtained with the AM/FM decomposition, intending to illustrate manipulations in the AM/FM domain that can be applied as interesting audio effects. We provide high-quality AM/FM effects and their implementations, alongside a brief objective evaluation. Audio samples and codes for real-time operation are also supplied.
Download Searching for Music Mixing Graphs: A Pruning Approach Music mixing is compositional — experts combine multiple audio processors to achieve a cohesive mix from dry source tracks. We propose a method to reverse engineer this process from the input and output audio. First, we create a mixing console that applies all available processors to every chain. Then, after the initial console parameter optimization, we alternate between removing redundant processors and fine-tuning. We achieve this through differentiable implementation of both processors and pruning. Consequently, we find a sparse mixing graph that achieves nearly identical matching quality of the full mixing console. We apply this procedure to drymix pairs from various datasets and collect graphs that also can be used to train neural networks for music mixing applications.
Download Automatic drum transcription with convolutional neural networks Automatic drum transcription (ADT) aims to detect drum events in polyphonic music. This task is part of the more general problem of transcribing a music signal in terms of its musical score and additionally can be very interesting for extracting high level information e.g. tempo, downbeat, measure. This article has the objective to investigate the use of Convolutional Neural Networks (CNN) in the context of ADT. Two different strategies are compared. First an approach based on a CNN based detection of drum only onsets is combined with an algorithm using Non-negative Matrix Deconvolution (NMD) for drum onset transcription. Then an approach relying entirely on CNN for the detection of individual drum instruments is described. The question of which loss function is the most adapted for this task is investigated together with the question of the optimal input structure. All algorithms are evaluated using the publicly available ENST Drum database, a widely used established reference dataset, allowing easy comparison with other algorithms. The comparison shows that the purely CNN based algorithm significantly outperforms the NMD based approach, and that the results are significantly better for the snare drum, but slightly worse for both the bass drum and the hi-hat when compared to the best results published so far and ones using also a neural network model.
Download Modelling Experts’ Decisions on Assigning Narrative Importances of Objects in a Radio Drama Mix There is an increasing number of consumers of broadcast audio who suffer from a degree of hearing impairment. One of the methods developed for tackling this issue consists of creating customizable object-based audio mixes where users can attenuate parts of the mix using a simple complexity parameter. The method relies on the mixing engineer classifying audio objects in the mix according to their narrative importance. This paper focuses on automating this process. Individual tracks are classified based on their music, speech, or sound effect content. Then the decisions for assigning narrative importance to each segment of a radio drama mix are modelled using mixture distributions. Finally, the learned decisions and resultant mixes are evaluated using the Short Term Objective Intelligibility, with reference to the narrative importance selections made by the original producer. This approach has applications for providing customizable mixes for legacy content, or automatically generated media content where the engineer is not able to intervene.
Download GPGPU Audio Benchmark Framework Acceleration of audio workloads on generally-programmable GPU (GPGPU) hardware offers potentially high speedup factors, but also presents challenges in terms of development and deployment. We can increasingly depend on such hardware being available in users’ systems, yet few real-time audio products use this resource. We propose a suite of benchmarks to qualify a GPU as suitable for batch or real-time audio processing. This includes both microbenchmarks and higher-level audio domain benchmarks. We choose metrics based on application, paying particularly close attention to latency tail distribution. We propose an extension to the benchmark framework to more accurately simulate the real-world request pattern and performance requirements when running in a digital audio workstation. We run these benchmarks on two common consumer-level platforms: a PC desktop with a recent midrange discrete GPU and a Macintosh desktop with unified CPUGPU memory architecture.
Download Time and Frequency Domain Room Compensation applied to Wave Field Synthesis In sound rendering systems using loudspeakers, the listening room adds echoes not considered by the reproduction system, thus deteriorating the rendered audio signal. Specifically, Wave Field Synthesis is a 3D audio reproduction system, which allows synthesizing a realistic sound field in a wide area by using arrays of loudspeakers. This paper proposes a room compensation approach based on a multichannel inverse filter bank calculated to compensate the room effects at selected points within the listening area. Time domain and frequency domain algorithms are proposed to accurately compute the bank of inverse filters. A comparative study between these algorithms by means of laboratory experiments is presented.
Download Autoencoding Neural Networks as Musical Audio Synthesizers A method for musical audio synthesis using autoencoding neural networks is proposed. The autoencoder is trained to compress and reconstruct magnitude short-time Fourier transform frames. The autoencoder produces a spectrogram by activating its smallest hidden layer, and a phase response is calculated using real-time phase gradient heap integration. Taking an inverse short-time Fourier transform produces the audio signal. Our algorithm is light-weight when compared to current state-of-the-art audio-producing machine learning algorithms. We outline our design process, produce metrics, and detail an open-source Python implementation of our model.