This package contains scripts to perform artificial bandwidth extension (ABE), super-wide bandwidth extension (SWBE) and some useful functions
Copyright (C) 2018 EURECOM, France. This work is licensed under the Creative Commons Attribution-NonCommercial-ShareAlike 4.0 International License. To view a copy of this license, visit http://creativecommons.org/licenses/by-nc-sa/4.0/
-
ABE_explicit_memory_ICASSP18
- This package contains scripts to perform ABE using explicit memory inclusion approach presented in the following publication: P. Bachhav, M. Todisco, and N. Evans, “Exploiting explicit memory inclusion for artificial bandwidth extension,” in Proc. of IEEE Int. Conf. on Acoustics, Speech and Signal Processing (ICASSP), 2018, pp. 5459-63, Calgary, Canada.
- Refer "Readme.txt" in folder "ABE_explicit_memory_ICASSP18" for further details
-
SWBE_LPAS_ICASSP18
- This package contains scripts to perform SWBE using linear prediction based analysis-synthesis approach presented in the following publication: P. Bachhav, M. Todisco, and N. Evans, “Efficient super-wide bandwidth extension using linear prediction based analysis synthesis,” in Proc. of IEEE Int. Conf. on Acoustics, Speech and Signal Processing (ICASSP), 2018, pp. 5429-5433, Calgary, Canada.
- Refer "Readme.txt" in folder "SWBE_LPAS_ICASSP18" for further details
-
ABE_SSAE_IS18
- This package contains scripts to perform ABE with memory inclusion using Semi-supervised Stacked Auto-encoders approach presented in the following publication: P. Bachhav, M. Todisco, and N. Evans, "Artificial Bandwidth Extension with Memory Inclusion using Semi-supervised Stacked Auto-encoders", in Proc. of INTERSPEECH, pp. 1185-89, Hyderabad, India.
- Refer "Readme.txt" in folder "ABE_SSAE_IS18" for further details
-
ABE_CVAE_ICASSP19
- This package contains scripts to perform ABE with latent features obtained using a deep generative model based approach presented in the following publication: P. Bachhav, M. Todisco, and N. Evans, "Latent Representation Learning for Artificial Bandwidth Extension using a Conditional Variational Auto-Encoder", accepted in ICASSP 2019.
- Refer "Readme.txt" in folder "ABE_CVAE_ICASSP19" for further details
-
utilities
- This folder contains MATLAB scripts or functions used for ABE and SWBE tasks listed above
-
Filters
- This folder contains impulse response of FIR filters used
-
Examples
- This folder contains usage examples of some functions available in folder "utilities"
All the scripts are successfully tested on MATLAB 2018a
For any query, please contact:
- Pramod Bachhav (bachhav at eurecom dot fr)
- Massimiliano Todisco (todisco at eurecom dot fr)
- Nicholas Evans (evans at eurecom dot fr)