37 citations · 88 across the 28 of their papers we have counts for
31 papers
The Fragility and Dynamics of LLM Pruning-Recovery Steps
Luis Frentzen Salim, Komang Rinartha, Chia-Chun Wu +3
Structural pruning followed by a recovery stage is the standard recipe for compressing large language models (LLMs), yet the two stages are typically designed and tuned in isolatio…
Rescuing Performance from the Demo: Co-Designing Drum Gesture Mappings with a Percussionist
Jordie Shier, Teresa Pelinski, Charalampos Saitis +2
Augmenting instruments with sensors and neural network mappings is a well-explored digital musical instrument design approach. While augmentations can create new expressive opportu…
EMORSION: Examining the Impact of Audio Parameters on Emotional Responses and Immersion in Film
Nelly Garcia, Ruby Crocker, Bleiz M Del Sette +4
EMORSION is an exploratory proof-of-concept study examining how film audio design shapes audience emotion and immersion in acinema setting. Four film scenes were selected across th…
Assessing the Alignment of Audio Representations with Timbre Similarity Ratings
Haokun Tian, Stefan Lattner, Charalampos Saitis
Psychoacoustical so-called "timbre spaces" map perceptual similarity ratings of instrument sounds onto low-dimensional embeddings via multidimensional scaling, but suffer from scal…
Audio synthesizer inversion in symmetric parameter spaces with approximately equivariant flow matching
Ben Hayes, Charalampos Saitis, György Fazekas
Many audio synthesizers can produce the same signal given different parameter configurations, meaning the inversion from sound to parameters is an inherently ill-posed problem. We…
Mamba-Diffusion Model with Learnable Wavelet for Controllable Symbolic Music Generation
Jincheng Zhang, György Fazekas, Charalampos Saitis
The recent surge in the popularity of diffusion models for image synthesis has attracted new attention to their potential for generation tasks in other domains. However, their appl…
Designing Neural Synthesizers for Low-Latency Interaction
Franco Caspe, Jordie Shier, Mark Sandler +2
Neural Audio Synthesis (NAS) models offer interactive musical control over high-quality, expressive audio generators. While these models can operate in real-time, they often suffer…
Hybrid Losses for Hierarchical Embedding Learning
Haokun Tian, Stefan Lattner, Brian McFee +1
In traditional supervised learning, the cross-entropy loss treats all incorrect predictions equally, ignoring the relevance or proximity of wrong labels to the correct answer. By l…
Foundation Models for Music: A Survey
Yinghao Ma, Anders Øland, Anton Ragni +39
In recent years, foundation models (FMs) such as large language models (LLMs) and latent diffusion models (LDMs) have profoundly impacted diverse sectors, including music. This com…
Automatic Detection of Moral Values in Music Lyrics
Vjosa Preniqi, Iacopo Ghinassi, Julia Ive +2
Moral values play a fundamental role in how we evaluate information, make decisions, and form judgements around important social issues. The possibility to extract morality rapidly…
Real-time Timbre Remapping with Differentiable DSP
Jordie Shier, Charalampos Saitis, Andrew Robertson +1
Timbre is a primary mode of expression in diverse musical contexts. However, prevalent audio-driven synthesis methods predominantly rely on pitch and loudness envelopes, effectivel…
MoralBERT: A Fine-Tuned Language Model for Capturing Moral Values in Social Discussions
Vjosa Preniqi, Iacopo Ghinassi, Julia Ive +2
Moral values play a fundamental role in how we evaluate information, make decisions, and form judgements around important social issues. Controversial topics, including vaccination…
Fast Diffusion-GAN Model for Generating Symbolic Music with Emotions
Jincheng Zhang, György Fazekas, Charalampos Saitis
Diffusion models have shown promising results for a wide range of generative tasks with continuous data, such as image and audio synthesis. However, little progress has been made o…
Composer Style-specific Symbolic Music Generation Using Vector Quantized Discrete Diffusion Models
Jincheng Zhang, György Fazekas, Charalampos Saitis
Emerging Denoising Diffusion Probabilistic Models (DDPM) have become increasingly utilised because of promising results they have achieved in diverse generative tasks with continuo…
Differentiable Modelling of Percussive Audio with Transient and Spectral Synthesis
Jordie Shier, Franco Caspe, Andrew Robertson +3
Differentiable digital signal processing (DDSP) techniques, including methods for audio synthesis, have gained attention in recent years and lend themselves to interpretability in…
A Review of Differentiable Digital Signal Processing for Music & Speech Synthesis
Ben Hayes, Jordie Shier, György Fazekas +2
The term "differentiable digital signal processing" describes a family of techniques in which loss function gradients are backpropagated through digital signal processors, facilita…
Interactive Neural Resonators
Rodrigo Diaz, Charalampos Saitis, Mark Sandler
In this work, we propose a method for the controllable synthesis of real-time contact sounds using neural resonators. Previous works have used physically inspired statistical metho…
The language of sounds unheard: Exploring musical timbre semantics of large language models
Kai Siedenburg, Charalampos Saitis
Semantic dimensions of sound have been playing a central role in understanding the nature of auditory sensory experience as well as the broader relation between perception, languag…
The Responsibility Problem in Neural Networks with Unordered Targets
Ben Hayes, Charalampos Saitis, György Fazekas
We discuss the discontinuities that arise when mapping unordered objects to neural network outputs of fixed permutation, referred to as the responsibility problem. Prior work has p…
Sinusoidal Frequency Estimation by Gradient Descent
Ben Hayes, Charalampos Saitis, György Fazekas
Sinusoidal parameter estimation is a fundamental task in applications from spectral analysis to time-series forecasting. Estimating the sinusoidal frequency parameter by gradient d…
Rigid-Body Sound Synthesis with Differentiable Modal Resonators
Rodrigo Diaz, Ben Hayes, Charalampos Saitis +2
Physical models of rigid bodies are used for sound synthesis in applications from virtual environments to music production. Traditional methods such as modal synthesis often rely o…
"More Than Words": Linking Music Preferences and Moral Values Through Lyrics
Vjosa Preniqi, Kyriaki Kalimeri, Charalampos Saitis
This study explores the association between music preferences and moral values by applying text analysis techniques to lyrics. Harvesting data from a Facebook-hosted application, w…
Deep Conditional Representation Learning for Drum Sample Retrieval by Vocalisation
Alejandro Delgado, Charalampos Saitis, Emmanouil Benetos +1
Imitating musical instruments with the human voice is an efficient way of communicating ideas between music producers, from sketching melody lines to clarifying desired sonorities.…
Deep Embeddings for Robust User-Based Amateur Vocal Percussion Classification
Alejandro Delgado, Emir Demirel, Vinod Subramanian +2
Vocal Percussion Transcription (VPT) is concerned with the automatic detection and classification of vocal percussion sound events, allowing music creators and producers to sketch…