Modèle physique variationnel pour l'estimation de réponses impulsionnelles de salles

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Lalay, Louis, Fontaine, Mathieu, Badeau, Roland
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866916838310412288
author Lalay, Louis
Fontaine, Mathieu
Badeau, Roland
author_facet Lalay, Louis
Fontaine, Mathieu
Badeau, Roland
contents Room impulse response estimation is essential for tasks like speech dereverberation, which improves automatic speech recognition. Most existing methods rely on either statistical signal processing or deep neural networks designed to replicate signal processing principles. However, combining statistical and physical modeling for RIR estimation remains largely unexplored. This paper proposes a novel approach integrating both aspects through a theoretically grounded model. The RIR is decomposed into interpretable parameters: white Gaussian noise filtered by a frequency-dependent exponential decay (e.g. modeling wall absorption) and an autoregressive filter (e.g. modeling microphone response). A variational free-energy cost function enables practical parameter estimation. As a proof of concept, we show that given dry and reverberant speech signals, the proposed method outperforms classical deconvolution in noisy environments, as validated by objective metrics.
format Preprint
id arxiv_https___arxiv_org_abs_2507_08051
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Modèle physique variationnel pour l'estimation de réponses impulsionnelles de salles
Lalay, Louis
Fontaine, Mathieu
Badeau, Roland
Sound
Audio and Speech Processing
Signal Processing
Classical Physics
Room impulse response estimation is essential for tasks like speech dereverberation, which improves automatic speech recognition. Most existing methods rely on either statistical signal processing or deep neural networks designed to replicate signal processing principles. However, combining statistical and physical modeling for RIR estimation remains largely unexplored. This paper proposes a novel approach integrating both aspects through a theoretically grounded model. The RIR is decomposed into interpretable parameters: white Gaussian noise filtered by a frequency-dependent exponential decay (e.g. modeling wall absorption) and an autoregressive filter (e.g. modeling microphone response). A variational free-energy cost function enables practical parameter estimation. As a proof of concept, we show that given dry and reverberant speech signals, the proposed method outperforms classical deconvolution in noisy environments, as validated by objective metrics.
title Modèle physique variationnel pour l'estimation de réponses impulsionnelles de salles
topic Sound
Audio and Speech Processing
Signal Processing
Classical Physics
url https://arxiv.org/abs/2507.08051