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 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 Arbitrary Polygon Oscillator: Generalizing Polygonal Synthesis to Arbitrary Shapes, Morphing, and Three-Dimensional Polyhedra Polygonal synthesis generates audio by traversing the perimeter of a polygon with a phasor; prior work uses a constant angular velocity, whereas the proposed system adopts constant arc-length (perimeter) velocity. Existing formulations operate on regular, parametrically defined polygons, producing smooth timbral transitions within a single family of shapes. This paper generalizes polygonal synthesis around a unified arc-length engine: vertex data of any origin feed the same DSP pipeline. First, we adapt the oscillator to accept arbitrary vertex configurations from an external buffer, opening the possibility for a broad class of closed polygons — regular, irregular, or star-shaped — to function as a waveform generator. Second, a hybrid interpolation algorithm enables smooth morphing between polygons with unequal vertex counts, passing through intermediate shapes that have no parametric description. Third, we extend the paradigm to three dimensions: a convex polyhedron rotated about three axes is sliced by a fixed horizontal plane, and the resulting cross-section yields a continuously variable polygon controlled by the solid's orientation. The system runs in RNBO (Cycling '74) with a geometry caching strategy that avoids per-sample recomputation. Antialiasing combines a four-point polyBLAMP correction derived from runtime Bézier tangents with adaptive oversampling, adapting the correction geometrically to general vertex configurations without per-shape analytical derivation.
Download Using the Distribution Derivative Method to Model Acoustic Musical Instrument Sounds with Polynomial AM-FM Sinusoids The oscillatory modes of musical instrument sounds are commonly modeled with time-varying sinusoids. Several estimation methods model quasi-stationary oscillations accurately, yielding a high-quality representation. However, nonstationary oscillations such as attack transients are still very challenging to model accurately. In this work, we propose to model musical instrument sounds with polynomial modulation sinusoids (PMS) estimated with the distribution derivative method (DDM). DDM gives accurate parameter estimations for PMS with arbitrary order, allowing great flexibility in modeling temporal changes inside analysis frames as amplitude and frequency modulations. We used 39 musical instrument sounds to compare DDM objectively against the standard (SM+) and an adaptive sinusoidal model (eaQHM) using time and frequency error measures. We showed that DDM captures more oscillatory energy than SM+ or eaQHM by modeling PMS more accurately. A MUSHRA listening test with 18 selected sounds confirmed that DDM has higher perceptual quality than both SM+ and eaQHM and that DDM is almost perceptually indistinguishable from the original sounds.
Download Winding Numbers and Monodromy of Vector Bundles over a Circular Buffer The Möbius strip is perhaps the most recognizable topological object of general knowledge. It can be described mathematically in various ways including the formalism of line bundles. In this paper we discuss the bundle idea in the context of digital processing over a circular buffer and show how the idea leads to a more general notion known as monodromy, which describes the effect of the space on traversing a circle once. In this formulation, the monodromy of the Möbius strip is an orientation inversion characterized by a change in sign. This in turns leads to the concept of the winding number, which describes how many windings it takes to return to the original state. We discuss variable monodromy and illustrate that the winding number is robust under this variation. This will allow us to interpret previous disparate results in audio signal processing from Möbius waveguides to chaotic oscillators in delay loops in one unified framework. We close by showing how extending from line to vector bundles opens up the notion of braids to describe monodromy.
Download A Clipping Prevention Method for All-Pass Digital Filters with Time-Varying Coefficients A clipping prevention method is proposed for first- and second-order all-pass filters with time-varying coefficients. Unlike conventional anti-clipping or declipping approaches, the method operates directly on the coefficient dynamics and does not rely on assumptions about internal energy evolution, by just asking that the input signal is not already clipping. The core idea is to control the deviation between the output of the time-varying filter and that of an equivalent static all-pass structure with constant coefficients. By adaptively limiting this deviation at runtime, the output is constrained below a prescribed clipping threshold (typically unit magnitude). The method is active only during short transients where clipping would occur, after which the coefficients are released to reach their target values. This preserves the integrity of the input signal and the numerical properties of the all-pass filter. Experimental results confirm the expected behavior even in scenarios where energy-preserving all-pass structures exceed the clipping threshold, suggesting the proposed approach as a practical solution for robust dynamic filter implementations with limited additional computational cost, suitable especially for embedded digital audio processing hardware.
Download A DDSP Framework for Adaptive Room Equalization Adaptive room equalization remains challenging under time-varying acoustic conditions and complex excitation signals, such as music. In these scenarios, classical filtered-x least mean squares (Fx-LMS) methods falter due to their rigid formulation. We present a modular differentiable digital signal processing (DDSP) framework for closed-loop adaptive room equalization that recovers Fx-LMS as a special case through automatic differentiation. The framework supports interchangeable EQ structures, response estimation methods, loss functions, and optimizers. Experiments with time-varying measured room impulse responses show that frequency-domain objectives provide more stable adaptation than time-domain objectives in the considered scenarios. Relative to the non-equalized response, system distance is reduced by 70% and mel-spectral distance by 13% (worst-case scenario). We further examine how online room response estimation accuracy and frame length affect the trade-off between responsiveness and convergence stability. Overall, the framework provides a unified open-source basis for exploring synergies between classical adaptive filtering and DDSP-based optimization.
Download Exploring Parallelism and Energy Efficiency in a Multistage Linear-Phase Octave Filter Bank This paper presents a high-performance and energy-aware implementation of a multistage linear-phase octave filter bank for edge system-on-chip (SoC) platforms. The algorithm relies on a cascade of stretched FIR filter stages and complementary band splitting to preserve linear phase across all outputs. While effective, mapping such structures to embedded multicore CPUs introduces significant challenges regarding state management, task synchronization, memory-traffic efficiency, and energy-aware execution. These issues are especially relevant in block-based edge-audio processing, where high throughput must be balanced against the power constraints of mobile and embedded devices. We derive a cache-friendly sequential realization using a blocked streaming schedule and compact circular state. Building on this, we propose a parallel design based on an OpenMP task pipeline with explicit dependencies to preserve the filter-bank semantics without fine-grained synchronization in the filtering tasks. Experimental results on an NVIDIA Jetson Orin Nano module show that the optimized sequential version already sustains more than 1.18 M samples/s, while the task-level pipeline reaches speedups above 4.5× for suitable block sizes. Furthermore, our analysis reveals a clear trade-off between throughput and power, showing that the most energy-efficient operating point does not necessarily coincide with maximum performance on multicore edge SoCs.
Download PolyADAA: Improving Aliasing Reduction in Memoryless Nonlinearities Using Lagrange Interpolation and Polynomial Approximation Reducing the aliasing of nonlinear functions is an important problem in digital signal processing. The introduction of the Antiderivative Antialiasing (ADAA) method brought many benefits and is an active area of research. The current bottleneck in terms of aliasing reduction is the initial conversion from discrete- to continuous-time, which was previously done by linear interpolation. In this paper we derive PolyADAA, a method for computing the ADAA output when this conversion is done using higher order Lagrange interpolation. To obtain a viable solution, the nonlinear function is approximated using Chebyshev polynomials, which enable the ADAA integral to be computed. The paper provides numerical examples to show the effectiveness of the approach and discusses the advantages of the method.
Download Alias-Free Oscillator Synchronization via Additive Synthesis Oscillator synchronization is a widely used sound-synthesis technique, but straightforward digital implementations suffer from aliasing artifacts. This paper presents an alias-free method for digital emulation of oscillator synchronization of arbitrary periodic waveforms based on additive synthesis. Starting from a finite set of Fourier-series coefficients representing a bandlimited free-running waveform, we derive linear spectral-resampling transforms that map these coefficients to those of the bandlimited synchronized waveform. Beyond conventional hard synchronization, the proposed approach also supports two additional soft-synchronization modes. To address the high computational complexity of the proposed method, we introduce HASY, a 6 mm² application-specific integrated circuit (ASIC) fabricated in 65 nm CMOS technology. HASY generates one 96 kHz, 24 bit alias-free synchronized waveform with up to 512 harmonics and computes the spectral-resampling transform within only five audio-sample periods.