Emilie d'Olne

Emilie d'Olne

PhD student in Speech and Audio Processing

Demonstration

Speech Enhancement in Distributed Microphone Arrays Using Polynomial Eigenvalue Decomposition

This demonstration provides listening examples of speech enhancement in distributed microphone arrays using a PEVD algorithm presented in this paper.

In all cases, results are compared against

  • the best and worst raw microphone signals
  • the best and worst delay-and-sum beamformers (DSB)
  • the best and worst minimum variance distortionless response (MVDR) beamformers
  • the PEVD applied to all 6 raw microphone signals
  • the proposed method applying PEVD to the 3 DSBs.

This page was generated using trackswitch.js in [1].

Experiment 1

Anechoic recordings

Diffuse white Gaussian noise at 0 dB SNR.


Reverberant recordings (T60=400 ms)

Diffuse white Gaussian noise at 0 dB SNR.


Experiment 2: Mis-steering

Speech-shaped white Gaussian noise at 0 dB SNR with 90° mis-steering.

References

[1]   N. Werner, S. Balke, F-R Stöter, M. Müller, B. Edler “trackswitch.js: A Versatile Web-Based Audio Player for Presenting Scientifc Results.” 3rd Web Audio Conference, London, UK. 2017.