
Only the Per-Channel Energy Normalisation (PCEN) layer of the LEArnable Front-end (LEAF) model learns during training, while the Gabor filterbank and Gaussian low-pass filters remain unchanged. Adapting the PCEN layer using a small amount of noisy data can improve the performance of a LEAF model trained on clean speech when deployed in noisy environments.


coremsg

investigating-the-learnable-components-of-the-learnable-front-end-leaf-and-adapting-its-per-channel-energy-normalisation-pcen-for-noisy-conditions


Investigating the Learnable Components of the LEArnable Front-end (LEAF) and Adapting its Per-Channel Energy Normalisation (PCEN) for Noisy Conditions


title_rewrite


The networks learn to denoise and dereverberate the microphone signals to better correlate them and consequently estimate the source position.


interpreting-end-to-end-deep-learning-models-for-speech-source-localization-using-layer-wise-relevance-propagation


Interpreting End-to-End Deep Learning Models for Speech Source Localization Using Layer-wise Relevance Propagation
