Download Sound Matching with a Differentiable Karplus-Strong Algorithm We present a self-supervised, event-based sound matching model using a differentiable extended Karplus-Strong algorithm. To avoid relying on external onset and fundamental frequency detectors, we explore training methodologies combining parameter losses on synthetic data with audio losses. We demonstrate that time-domain fractional delay interpolation provides gradient accuracy comparable to frequency-sampling while avoiding time-aliasing in highly resonant time-varying scenarios. Through systematic gradient analysis, we reveal that standard spectral losses provide no meaningful directional gradients for onset times, heavily degrading joint training. Training exclusively with parameter losses on synthetic data effectively learns fundamental frequency, timbral parameters, and onset times, but struggles to generalise to monophonic studio recordings of plucked guitar. External detectors combined with audio losses generalise best, isolating the model to timbre optimisation. While our Karplus-Strong decoder recovers interpretable parameters and naturally captures the transient characteristics of plucked guitar, Harmonics plus Noise baselines yield higher reconstruction fidelity by most metrics.
Download FM Synthesizer Audio-Parameter Shared Embeddings Given a target sound, finding the synthesizer preset that best reproduces it remains a core problem in sound design. Existing methods treat synthesis parameters as flat vectors, discarding the signal routing and parameter interactions that produce audio. We make two contributions. First, to learn a representation of parameters including their signal routing, we design a graph neural network whose message passing structure imitates FM signal processing. Second, we adapt the multimodal objective from SLAP to learn joint embeddings of audio and FM synthesizer parameters, enabling preset retrieval from a gallery. We focus on the Yamaha DX7, where six identical sinusoid operators interact according to one of 32 routing topologies. Our graph encoder's message passing weights are shared across all nodes and layers, enabling processing of arbitrary topologies of any size. When every topology is seen during training, the DX7-GNN and two baselines achieve strong audio-to-preset retrieval. When some topologies are held out for testing, the DX7-GNN substantially outperforms both baselines despite having the fewest parameters. Our ablations further support the claim that imitating FM signal flow in a parameter encoder improves generalization to unseen topologies.
Download Fast Parametric Matrices for Lossless Feedback Delay Networks This paper presents a framework for designing creative reverbs using parametric orthogonal feedback matrices on Feedback Delay Networks (FDNs) through recursive Kronecker products of 2D rotation and reflection matrices. By parameterizing each 2×2 kernel with a single angle, we construct a family of 2M×2M orthogonal matrices that maintain losslessness while enabling continuous control over network topology. We then exploit their recursive definition to compute the feedback operation with an O(N log₂ N) divide-and-conquer algorithm that matches the Fast Walsh-Hadamard Transform time complexity while offering parametric flexibility. Strategic manipulation of individual kernel angles enables creative sound design applications, such as stereo cross-coupling, selective freeze, and time-varying modulation for resonance breaking.