Skip to main content

Open Access The Master Hearing Aid: A PC-Based Platform for Algorithm Development and Evaluation

Development and evaluation of algorithms for digital signal processing in hearing aids includes many stages from the first implementation of the algorithmic idea, technical evaluations, subjective evaluations with patients up to field tests, and involves several expert groups, usually physicists, engineers, audiologists and hearing aid acousticians. In order to facilitate this complex process, a common platform for development and evaluation is desirable that covers the whole development process and integrates seamlessly the work of the different expert groups. This paper discusses the possibility of using PC-based hardware for this task. Considering the Master-Hearing-Aid (MHA) developed within the HörTech center of competence on hearing technology as an example, it is shown that the approach of using standardized hardware and software is most promising. The MHA allows for the integration of algorithm development using standard software like Matlab and algorithm evaluation using low-delay real-time processing in combination with user-friendly graphical control interfaces. Performance tests show that real-time processing on modern PC-based hardware is possible even with algorithms written in the Matlab script language, although in this case processing delay is larger and floating point performance is smaller than in algorithms programmed in the C language. For waveform processing, a total delay of 4.35 ms can be reached at less than 5% CPU-load when implemented in C, and a delay of about 94 ms at about 20% CPU-load when implemented as a Matlab script. For spectral FFT-based processing, a total delay of 7.25 ms can be reached at less than 10% CPU-load when implemented in C, and a delay of about 187 ms at about 30% CPU-load when implemented as a Matlab script.

Document Type: Research Article

Publication date: 01 July 2006

  • Access Key
  • Free content
  • Partial Free content
  • New content
  • Open access content
  • Partial Open access content
  • Subscribed content
  • Partial Subscribed content
  • Free trial content