• English
  • Deutsch
  • Log In
    Password Login
    Research Outputs
    Fundings & Projects
    Researchers
    Institutes
    Statistics
Repository logo
Fraunhofer-Gesellschaft
  1. Home
  2. Fraunhofer-Gesellschaft
  3. Scopus
  4. Blind Reverberation Time Estimation in Dynamic Acoustic Conditions
 
  • Details
  • Full
Options
2022
Conference Paper
Title

Blind Reverberation Time Estimation in Dynamic Acoustic Conditions

Abstract
The estimation of reverberation time from real-world signals plays a central role in a wide range of applications. In many scenarios, acoustic conditions change over time which in turn requires the estimate to be updated continuously. Previously proposed methods involving deep neural networks were mostly designed and tested under the assumption of static acoustic conditions. In this work, we show that these approaches can perform poorly in dynamically evolving acoustic environments. Motivated by a recent trend towards data-centric approaches in machine learning, we propose a novel way of generating training data and demonstrate, using an existing deep neural network architecture, the considerable improvement in the ability to follow temporal changes in reverberation time.
Author(s)
Götz, Philipp
International Audio Laboratories Erlangen
Tuna, Cagdas
Fraunhofer-Institut für Integrierte Schaltungen IIS  
Walther, Andreas  
Fraunhofer-Institut für Integrierte Schaltungen IIS  
Habets, Emanuël A.P.
International Audio Laboratories Erlangen
Mainwork
IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2022. Proceedings  
Conference
International Conference on Acoustics, Speech, and Signal Processing 2022  
Open Access
DOI
10.1109/ICASSP43922.2022.9746457
Language
English
Fraunhofer-Institut für Integrierte Schaltungen IIS  
Keyword(s)
  • convolutional recurrent neural networks

  • Data-centric AI

  • Dynamic acoustic conditions

  • reverberation time estimation

  • Cookie settings
  • Imprint
  • Privacy policy
  • Api
  • Contact
© 2024