This paper introduces scattering transform for speech emotion recognition (SER). Scattering
transform generates feature representations which remain stable to deformations and …

In this article, we investigate the notion of model-based deep learning in the realm of music
information research (MIR). Loosely speaking, we refer to the term model-based deep …

Commonly used features in spoken language identification (LID), such as mel-spectrogram
or MFCC, lose high-frequency information due to windowing. The loss further increases for …

Sound matching algorithms seek to approximate a target waveform by parametric audio
synthesis. Deep neural networks have achieved promising results in matching sustained …

Joint time-frequency scattering (JTFS) is a convolutional operator in the time-frequency
domain which extracts spectrotemporal modulations at various rates and scales. It offers an …

Sound retrieval for assisted music composition depends on evaluating similarity between
musical instrument sounds, which is partly influenced by playing techniques. Previous …

Deep convolutional networks (convnets) in the time–frequency domain can learn an
accurate and fine-grained categorization of sounds. For example, in the context of music …

Perceptual sound matching (PSM) aims to find the input parameters to a synthesizer so as to
best imitate an audio target. Deep learning for PSM optimizes a neural network to analyze …

The characterization of the musical timbre, which allows the quantitative evaluation of
audios, is still an open-ended research topic. This paper evaluates a set of dimensionless …

Microcanonical gradient descent is a sampling procedure for energy-based models allowing
for efficient sampling of distributions in high dimension. It works by transporting samples …