Search

Electric guitar tone modeling typically focuses on the non-linear transformation from clean to amplifier-rendered audio. Traditional methods rely on one-to-one mappings, incorporating device parameters into neural models to replicate specific amplifiers. However, these methods are limited by the need for specific training data. In this paper, we adapt a model based on the previous work, which leverages a tone embedding encoder and a feature wise linear modulation (FiLM) condition method. In this work, we altered conditioning method using a hypernetwork-based gated convolutional network (GCN) to generate audio that blends clean input with the tone characteristics of reference audio. By extending the training data to cover a wider variety of amplifier tones, our model is able to capture a broader range of tones. Additionally, we developed a real-time plugin to demonstrate the system's practical application, allowing users to experience its performance interactively. Our results indicate that the proposed system achieves superior tone modeling versatility compared to traditional methods., Comment: demo of the ISMIR paper

Neural network models for guitar amplifier emulation, while being effective, often demand high computational cost and lack interpretability. Drawing ideas from physical amplifier design, this paper aims to address these issues with a new differentiable digital signal processing (DDSP)-based model, called ``DDSP guitar amp,'' that models the four components of a guitar amp (i.e., preamp, tone stack, power amp, and output transformer) using specific DSP-inspired designs. With a set of time- and frequency-domain metrics, we demonstrate that DDSP guitar amp achieves performance comparable with that of black-box baselines while requiring less than 10\% of the computational operations per audio sample, thereby holding greater potential for usages in real-time applications., Comment: Preprint paper

Replicating analog device circuits through neural audio effect modeling has garnered increasing interest in recent years. Existing work has predominantly focused on a one-to-one emulation strategy, modeling specific devices individually. In this paper, we tackle the less-explored scenario of one-to-many emulation, utilizing conditioning mechanisms to emulate multiple guitar amplifiers through a single neural model. For condition representation, we use contrastive learning to build a tone embedding encoder that extracts style-related features of various amplifiers, leveraging a dataset of comprehensive amplifier settings. Targeting zero-shot application scenarios, we also examine various strategies for tone embedding representation, evaluating referenced tone embedding against two retrieval-based embedding methods for amplifiers unseen in the training time. Our findings showcase the efficacy and potential of the proposed methods in achieving versatile one-to-many amplifier modeling, contributing a foundational step towards zero-shot audio modeling applications., Comment: ISMIR 2024

Let $A$ be a limsup random fractal with indices $\gamma_1, ~\gamma_2 ~$and $\delta$ on $[0,1]^d$. We determine the hitting probability $\mathbb{P}(A\cap G)$ for any analytic set $G$ with the condition $(\star)$$\colon$ $\dim_{\rm H}(G)>\gamma_2+\delta$, where $\dim_{\rm H}$ denotes the Hausdorff dimension. This extends the correspondence of Khoshnevisan, Peres and Xiao [10] by relaxing the condition that the probability $P_n$ of choosing each dyadic hyper-cube is homogeneous and $\lim\limits_{n\to\infty}\frac{\log_2P_n}{n}$ exists. We also present some counterexamples to show the Hausdorff dimension in condition $(\star)$ can not be replaced by the packing dimension., Comment: 12 pages

The goal of this paper is to define and investigate those topological pressures, which is an extension of topological entropy presented by Feng and Huang [13], of continuous transformations. This study reveals the similarity between many known results of topological pressure. More precisely, the investigation of the variational principle is given and related propositions are also described. That is, this paper defines the measure theoretic pressure $P_{\mu}(T,f)$ for any $\mu\in{\mathcal M(X)}$, and shows that $P_B(T,f,K)=\sup\bigr\{P_{\mu}(T,f):{\mu}\in{\mathcalM(X)},{\mu}(K)=1\bigr\}$, where $K\subseteq X$ is a non-empty compact subset and $P_B(T,f,K)$ is the Bowen topological pressure on $K$. Furthermore, if $Z\subseteq X$ is an analytic subset, then $P_B(T,f,Z)=\sup\bigr\{P_B(T,f,K):K\subseteq Z\ \text{is compact}\bigr\}$. However, this analysis relies on more techniques of ergodic theory and topological dynamics., Comment: 15 pages. arXiv admin note: substantial text overlap with arXiv:1012.1103, arXiv:1111.7121 by other authors

This paper defines the pressure for asymptotically subadditive potentials under a mistake function, including the measuretheoretical and the topological versions. Using the advanced techniques of ergodic theory and topological dynamics, we reveals a variational principle for the new defined topological pressure without any additional conditions on the potentials and the compact metric space., Comment: 13pages