Noise reduction using neural lateral inhibition for speech enhancement

Research output: Contribution to journalArticlepeer-review

49 Downloads (Pure)

Abstract

Recurrent spiking neurons with lateral inhibition connection play a vital role in human’s brain functional abilities. In this paper, we propose a novel noise reduction method that is based on neuron rate coding and bio-inspired spiking neural network architecture. The excitatory-inhibitory topology in the network acts as the temporal characteristic synchrony and coincidence detector that removes uncorrelated noisy spikes. A LIF source encoder is introduced along with the network. The network uses generated binary Short-Time Fourier Transform (STFT) masks according to the rate of processed spike train, which is used to reconstruct the denoised speech signal. The technique is evaluated on noisy speech samples with 5 types of real-world additive noise with different noise strength.
Original languageEnglish
Number of pages6
Journal International Journal of Machine Learning and Computing
Publication statusAccepted/In press - 10 Oct 2019

Keywords

  • spiking
  • neural network
  • speech enhancement
  • noise reduction
  • lateral inhibition

Fingerprint

Dive into the research topics of 'Noise reduction using neural lateral inhibition for speech enhancement'. Together they form a unique fingerprint.

Cite this