Speech Modeling with a Hierarchical Transformer Dynamical VAE - Inria - Institut national de recherche en sciences et technologies du numérique Accéder directement au contenu
Communication Dans Un Congrès Année : 2023

Speech Modeling with a Hierarchical Transformer Dynamical VAE

Résumé

The dynamical variational autoencoders (DVAEs) are a family of latent-variable deep generative models that extends the VAE to model a sequence of observed data and a corresponding sequence of latent vectors. In almost all the DVAEs of the literature, the temporal dependencies within each sequence and across the two sequences are modeled with recurrent neural networks. In this paper, we propose to model speech signals with the Hierarchical Transformer DVAE (HiT-DVAE), which is a DVAE with two levels of latent variable (sequence-wise and frame-wise) and in which the temporal dependencies are implemented with the Transformer architecture. We show that HiT-DVAE outperforms several other DVAEs for speech spectrogram modeling, while enabling a simpler training procedure, revealing its high potential for downstream low-level speech processing tasks such as speech enhancement.

Dates et versions

hal-04132313 , version 1 (19-06-2023)

Licence

Paternité

Identifiants

Citer

Xiaoyu Lin, Xiaoyu Bie, Simon Leglaive, Laurent Girin, Xavier Alameda-Pineda. Speech Modeling with a Hierarchical Transformer Dynamical VAE. ICASSP 2023 - IEEE International Conference on Acoustics, Speech and Signal Processing, Jun 2023, Rhodes, Greece. pp.1-5, ⟨10.1109/ICASSP49357.2023.10096751⟩. ⟨hal-04132313⟩
64 Consultations
0 Téléchargements

Altmetric

Partager

Gmail Facebook X LinkedIn More