Neural network for multi-exponential sound energy decay analysis

Abstract

An established model for sound energy decay functions (EDFs) is the superposition of multiple exponentials and a noise term. This work proposes a neural-network-based approach for estimating the model parameters from EDFs. The network is trained on synthetic EDFs and evaluated on two large datasets of over 20000 EDF measurements conducted in various acoustic environments. The evaluation shows that the proposed neural network architecture robustly estimates the model parameters from large datasets of measured EDFs, while being lightweight and computationally efficient. An implementation of the proposed neural network is publicly available.

arXiv commentThe following article has been submitted to the Journal of the Acoustical Society of America (JASA). After it is published, it will be found at http://asa.scitation.org/journal/jas

References in corpus (10)

• Adam: A Method for Stochastic Optimization
• Decoupled Weight Decay Regularization
• SGDR: Stochastic Gradient Descent with Warm Restarts
• Recent Trends in Deep Learning Based Natural Language Processing
• Deep Learning for Audio Signal Processing
• Machine learning in acoustics: theory and applications
• A Survey of Sound Source Localization with Deep Learning Methods
• Integrated perception with recurrent multi-task neural networks
• Fast acoustic scattering using convolutional neural networks
• Prediction of Object Geometry from Acoustic Scattering Using Convolutional Neural Networks