GLA-Grad: A Griffin-Lim Extended Waveform Generation Diffusion Model - Archive ouverte HAL Access content directly
Conference Papers Year : 2024

GLA-Grad: A Griffin-Lim Extended Waveform Generation Diffusion Model

Abstract

Diffusion models are receiving a growing interest for a variety of signal generation tasks such as speech or music synthesis. WaveGrad, for example, is a successful diffusion model that conditionally uses the mel spectrogram to guide a diffusion process for the generation of high-fidelity audio. However, such models face important challenges concerning the noise diffusion process for training and inference, and they have difficulty generating high-quality speech for speakers that were not seen during training. With the aim of minimizing the conditioning error and increasing the efficiency of the noise diffusion process, we propose in this paper a new scheme called GLA-Grad, which consists in introducing a phase recovery algorithm such as the Griffin-Lim algorithm (GLA) at each step of the regular diffusion process. Furthermore, it can be directly applied to an already-trained waveform generation model, without additional training or fine-tuning. We show that our algorithm outperforms state-of-the-art diffusion models for speech generation, especially when generating speech for a previously unseen target speaker.
Fichier principal
Vignette du fichier
ICASSP_2024_GLA_Grad___Preprint.pdf (354.4 Ko) Télécharger le fichier
Origin : Files produced by the author(s)

Dates and versions

hal-04424100 , version 1 (09-02-2024)

Identifiers

  • HAL Id : hal-04424100 , version 1

Cite

Haocheng Liu, Teysir Baoueb, Mathieu Fontaine, Jonathan Le Roux, Gael Richard. GLA-Grad: A Griffin-Lim Extended Waveform Generation Diffusion Model. IEEE International Conference on Acoustics, Speech and Signal Processing, Apr 2024, Seoul (Korea), South Korea. ⟨hal-04424100⟩
23 View
6 Download

Share

Gmail Facebook X LinkedIn More