MUS423 Research Seminars

Abstract: We propose a new modal sound synthesis method that rapidly estimates all acoustic transfer fields of a linear modal vibration model, and greatly reduces preprocessing costs. Instead of performing a separate frequency-domain Helmholtz radiation analysis for each mode, our method partitions vibration modes into chords using optimal mode conflation, then performs a single time-domain wave simulation for each chord. We then perform transfer deconflation on each chord’s time-domain radiation field using a specialized QR solver, and thereby extract the frequency-domain transfer functions of each mode. The precomputed transfer functions are represented for fast far-field evaluation, e.g., using multipole expansions.

Abstract: Classical DSP techniques have recently been overshadowed by deep learning for applications such as image recognition and audio generation. End-2-end learning has been key to this shift, enabling the optimization of high-dimensional nonlinear functions. However DSP techniques provide interpretability, modularity, and efficiency lacking from black-box deep networks. In this talk, I'll review the Differentiable Digital Signal Processing (DDSP) library, which enables direct integration of classic signal processing elements with deep learning methods.

Abstract: Conventional audio synthesis (TTS, voice conversion, enhancement, etc) often relies on acoustic feature representations (spectrogram, MFCC, F0, etc.) and a signal processing procedure that infers the waveform from these features. However, such procedures often introduce artifacts caused by insufficient information in the feature representation (e.g. iSTFT without the correct phase info) and/or an oversimplified synthesis process (e.g. a source-filter model). Recent advancements battle this problem using deep learning: WaveNet, for example, generates the waveform sample-by-sample based on acoustic features and previously generated samples using a dilated convolutional net.

Carmine Emanuele Cella, assistant professor in music and technology at CNMAT, will present work done in the last years in searching good signal representations that permit high-level manipulation of musical concepts. After the definition of a geometric approach to signal representation, the theory of sound-types and its application to music will be presented.  Finally, recent research on assisted orchestration will be shown and some possible musical applications will be proposed, with connections to deep learning methods.

CNMAT, University of California, Berkeley: http://www.carminecella.com/index.html