This repository contains the inference code and trained models for subband-LSTM-based noise PSD estimator, the algorithm is described in the papers:
- Xiaofei Li, Simon Leglaive, Laurent Girin, Radu Horaud. Audio-noise Power Spectral Density Estimation Using Long Short-term Memory. IEEE Signal Processing Letters, 2019.
Author: Xiaofei Li, Westlake University, China and INRIA Grenoble Rhone-Alpes, France