Hyperbolic Embeddings for Order-Aware Classification of Audio Effect Chains

2025-07-28

AI Summary

Effect processor ordering in audio effect chains critically shapes timbral output, yet existing research predominantly focuses on estimating effect types and parameters from wet signals, neglecting the fundamental problem of ordering identification. This paper introduces hyperbolic space embedding to effect chain recognition for the first time, proposing a neural network-based method that leverages hyperbolic geometry to model ordered effect chains. Exploiting the exponential growth property of hyperbolic space, our approach naturally encodes the non-commutativity and combinatorial explosion inherent in effect sequencing, while explicitly representing ordered chains as tree structures. Experiments on a guitar tone dataset demonstrate that, under optimized curvature, our method achieves significantly higher joint accuracy in identifying both effect types and their ordering compared to Euclidean-space baselines—effectively overcoming the longstanding limitation of conventional approaches that disregard sequential dependencies.

Technology Category

• Machine Learning: Matrix & Tensor Methods
• Cognitive Modeling & Cognitive Systems: Affective Computing
• Knowledge Representation and Reasoning: Geometric, Spatial, and Temporal Reasoning

Application Category

• Graph Algorithms and Modeling for the Web: Efficient manipulation of static and dynamic Web-related graphs
• Web Mining and Content Analysis: Robustness and generalizability of Web mining methods
• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs

Abstract

Audio effects (AFXs) are essential tools in music production, frequently applied in chains to shape timbre and dynamics. The order of AFXs in a chain plays a crucial role in determining the final sound, particularly when non-linear (e.g., distortion) or time-variant (e.g., chorus) processors are involved. Despite its importance, most AFX-related studies have primarily focused on estimating effect types and their parameters from a wet signal. To address this gap, we formulate AFX chain recognition as the task of jointly estimating AFX types and their order from a wet signal. We propose a neural-network-based method that embeds wet signals into a hyperbolic space and classifies their AFX chains. Hyperbolic space can represent tree-structured data more efficiently than Euclidean space due to its exponential expansion property. Since AFX chains can be represented as trees, with AFXs as nodes and edges encoding effect order, hyperbolic space is well-suited for modeling the exponentially growing and non-commutative nature of ordered AFX combinations, where changes in effect order can result in different final sounds. Experiments using guitar sounds demonstrate that, with an appropriate curvature, the proposed method outperforms its Euclidean counterpart. Further analysis based on AFX type and chain length highlights the effectiveness of the proposed method in capturing AFX order.

Problem

• Estimating audio effect types and order from wet signals
• Modeling non-commutative AFX chains using hyperbolic embeddings
• Improving classification accuracy for ordered effect combinations

Innovation

• Hyperbolic embeddings model AFX chains efficiently
• Neural network classifies ordered AFX combinations
• Tree-structured data represented in hyperbolic space