Modeling plate and spring reverberation using a DSP-informed deep nerual network

View the Project on GitHub mchijmma/modeling-plate-spring-reverb

Audio examples for the paper:

Martínez Ramírez M. A., Benetos, E. and Reiss J. D., “Modeling plate and spring reverberation using a DSP-informed deep neural network” submitted to the IEEE International Conference on Acoustics, Speech, and Signal Processing (ICASSP), Barcelona, Spain, May 2020.

plate

- input
- target
- model-1
- model-2

 

spring

- input
- target
- model-1
- model-2

 

Model


Block diagram of the proposed model; adaptive front-end, latent-space and synthesis back-end:


Detailed architecture of adaptive front-end:

Input frame size of 4096 samples and ±4 context frames.


Detailed architecture of the latent-space:


Detailed architecture of synthesis back-end:

Output frame size of 4096 samples.

 

Citation

@inproceedings{martinez2020modeling,
title={Modeling plate and spring reverberation using a DSP-informed deep neural network},
author={Mart'{i}nez Ram'{i}rez, Marco A and Benetos, Emmanouil and Reiss, Joshua D},
booktitle={arXiv preprint arXiv:1910.10105}
}