English

Zero-Shot Unsupervised and Text-Based Audio Editing Using DDPM Inversion

Sound 2024-05-30 v4 Machine Learning Audio and Speech Processing

Abstract

Editing signals using large pre-trained models, in a zero-shot manner, has recently seen rapid advancements in the image domain. However, this wave has yet to reach the audio domain. In this paper, we explore two zero-shot editing techniques for audio signals, which use DDPM inversion with pre-trained diffusion models. The first, which we coin ZEro-shot Text-based Audio (ZETA) editing, is adopted from the image domain. The second, named ZEro-shot UnSupervized (ZEUS) editing, is a novel approach for discovering semantically meaningful editing directions without supervision. When applied to music signals, this method exposes a range of musically interesting modifications, from controlling the participation of specific instruments to improvisations on the melody. Samples and code can be found in https://hilamanor.github.io/AudioEditing/ .

Cite

@article{arxiv.2402.10009,
  title  = {Zero-Shot Unsupervised and Text-Based Audio Editing Using DDPM Inversion},
  author = {Hila Manor and Tomer Michaeli},
  journal= {arXiv preprint arXiv:2402.10009},
  year   = {2024}
}

Comments

Accepted for ICML 2024; Examples and code available in https://hilamanor.github.io/AudioEditing/

R2 v1 2026-06-28T14:49:40.960Z