Signal Processing and Speech Communication Laboratory
hometheses & projects › Using sEMG for Disordered Speech Enhancement III -- Adaptive Threshold

Using sEMG for Disordered Speech Enhancement III -- Adaptive Threshold

Status
Finished
Type
Bachelor Project
Announcement date
10 Mar 2015
Student
Andreas Fuchs
Mentors
Research Areas

Short description

An electro-larynx is a device which is used to re-obtain speech after removal of the larynx. It is a small hand-held, battery-driven device and its resulting speech suffer from unnaturalness.

It turned out that surface electromyography (sEMG), i.d. muscle activity at the neck, is a possibility to control the on/off signal of the device in an automatic way without occupying one hand. In previous work we developed a hardware to capture sEMG signal and implemented a real-time system to operate the EL device in a hands-free way [1]. The threshold to turn on and off the EL device depends on the SNR of the EMG signal which is, in turn, speaker dependent.

Your Tasks

  • Review of literature
  • MATLAB implementation of adaptive threshold to operate the EL device
  • Documentation

Your Profile/Prerequisites

  • Motivation and interest in the topic
  • Background information in Matlab and (speech) signal processing

Contact:

Martin Hagmüller ({hagmueller}@tugraz.at or 0316/873 4377) Anna Katharina Fuchs ({anna.fuchs}@tugraz.at or 0316/873 4367)

References:

[1]A. K. Fuchs, C. Amon, and M. Hagmüller, “Speech/non-speech detection for electro-larynx speech using emg,” in Biosignals - 8th International Conference on Bio-Inspired and Signal Processing (BIOSIGNALS 2015), 2015, pp. 138 – 144.