Music and Cochlear Implants

Abstract logo

MusIC: Making Music More Accessible for Cochlear Implant Users

  • Enhancing the signing voice of music to improve music appreciation for CI users
  • Real-time music source separation to apply enhancement on singing voice
  • Parametrization of CI sound coding strategies for singing music
  • Sound Coding Strategy for music

Music Source Separation Using Deep Neural Networks:

Music samples of the Real-time Music Source Separation experiment. In this experiment a multilayer perceptron has been used to separate the singing voice from the instrumental accompaniment to remix the music track. It has been shown that CI users enjoy music more, when the singing voice is enhanced with respect to the background instruments. Our results indicate that CI users enjoy music more when the singing voice is 8 dB enhanced with respect to the background accompaniment.

Audio Example:

Below is a demo of the music tracks used in this experiment.

Music Track Example 1 Example 2
Original Mixture
Enhanced Mixture
Vocals Signal
Instruments Signal


  • Difficult to enjoy complex music as i.e. classical pieces for CI-users.
  • Investigating, which musical component (i.e. melody) in instrumental classical music CI-users want to enhance.
  • Testing, if selective attention decoding is possible for CI-users.
  • Feeding EEG-data into neural networks to enhance source separation of attended sound source.

Team Members


Head of Research Group:
Prof. Dr.-Ing. Waldo Nogueira

DHZ-Deutsches HörZentrum Hannover
Karl-Wiechert-Allee 3
30625 Hannover
Phone: +49 (0)511 532 8025
Fax: +49 (0)511 532 6833

About The VIANNA research team