We gratefully acknowledge support from
the Simons Foundation and member institutions.
Full-text links:

Download:

Current browse context:

cs.SD

Change to browse by:

References & Citations

DBLP - CS Bibliography

Bookmark

(what is this?)
CiteULike logo BibSonomy logo Mendeley logo del.icio.us logo Digg logo Reddit logo ScienceWISE logo

Computer Science > Sound

Title: Towards Lightweight Controllable Audio Synthesis with Conditional Implicit Neural Representations

Abstract: The high temporal resolution of audio and our perceptual sensitivity to small irregularities in waveforms make synthesizing at high sampling rates a complex and computationally intensive task, prohibiting real-time, controllable synthesis within many approaches. In this work we aim to shed light on the potential of Conditional Implicit Neural Representations (CINRs) as lightweight backbones in generative frameworks for audio synthesis.
Implicit neural representations (INRs) are neural networks used to approximate low-dimensional functions, trained to represent a single geometric object by mapping input coordinates to structural information at input locations. In contrast with other neural methods for representing geometric objects, the memory required to parameterize the object is independent of resolution, and only scales with its complexity. A corollary of this is that INRs have infinite resolution, as they can be sampled at arbitrary resolutions. To apply the concept of INRs in the generative domain we frame generative modelling as learning a distribution of continuous functions. This can be achieved by introducing conditioning methods to INRs.
Our experiments show that Periodic Conditional INRs (PCINRs) learn faster and generally produce quantitatively better audio reconstructions than Transposed Convolutional Neural Networks with equal parameter counts. However, their performance is very sensitive to activation scaling hyperparameters. When learning to represent more uniform sets, PCINRs tend to introduce artificial high-frequency components in reconstructions. We validate this noise can be minimized by applying standard weight regularization during training or decreasing the compositional depth of PCINRs, and suggest directions for future research.
Comments: Accepted to "Deep Generative Models and Downstream Applications" and "Machine Learning for Creativity and Design" workshops at NeurIPS 2021
Subjects: Sound (cs.SD); Machine Learning (cs.LG)
Cite as: arXiv:2111.08462 [cs.SD]
  (or arXiv:2111.08462v1 [cs.SD] for this version)

Submission history

From: Jan Zuiderveld [view email]
[v1] Sun, 14 Nov 2021 13:36:18 GMT (15619kb,D)
[v2] Thu, 2 Dec 2021 11:16:29 GMT (23944kb,D)

Link back to: arXiv, form interface, contact.