A weighted-variance variational autoencoder model for speech enhancement - Inria - Institut national de recherche en sciences et technologies du numérique Accéder directement au contenu
Communication Dans Un Congrès Année : 2024

A weighted-variance variational autoencoder model for speech enhancement

Résumé

We address speech enhancement based on variational autoencoders, which involves learning a speech prior distribution in the time-frequency (TF) domain. A zero-mean complex-valued Gaussian distribution is usually assumed for the generative model, where the speech information is encoded in the variance as a function of a latent variable. In contrast to this commonly used approach, we propose a weighted variance generative model, where the contribution of each spectrogram time-frame in parameter learning is weighted. We impose a Gamma prior distribution on the weights, which would effectively lead to a Student's t-distribution instead of Gaussian for speech generative modeling. We develop efficient training and speech enhancement algorithms based on the proposed generative model. Our experimental results on spectrogram auto-encoding and speech enhancement demonstrate the effectiveness and robustness of the proposed approach compared to the standard unweighted variance model.
Fichier principal
Vignette du fichier
St_VAE_ICASSP2024.pdf (256.33 Ko) Télécharger le fichier
Origine : Fichiers produits par l'(les) auteur(s)

Dates et versions

hal-03833827 , version 1 (28-10-2022)
hal-03833827 , version 2 (20-09-2023)

Licence

Paternité

Identifiants

Citer

Ali Golmakani, Mostafa Sadeghi, Xavier Alameda-Pineda, Romain Serizel. A weighted-variance variational autoencoder model for speech enhancement. ICASSP 2024 - International Conference on Acoustics Speech and Signal Processing, IEEE, Apr 2024, Seoul (Korea), South Korea. pp.1-5. ⟨hal-03833827v2⟩
195 Consultations
74 Téléchargements

Altmetric

Partager

Gmail Facebook X LinkedIn More