Open tools for the development and evaluation of hearing devices
There is a wealth of knowledge about how to adjust our hearing, but most of it is locked up in commercial products. Recently the US National Institutes of Health sponsored a research program to develop a suite of hardware and software that are open source. By this means they hope to create a new research ecosystem that can build and test improvements to the current paradigms. And of course the open source ecosystem will make it easier to put new products and ideas into the field.
Hendrik Kayser is part of the team from Oldeburg University developing reference software and hardware for an open hearing aid. He will be presenting his work at a special session on "Open Source Audio Processing Tools for Hearing Research" at the ASA meeting in San Diego the week before.
Who: Hendrik Kayser (Oldenburg)
What: Open tools for the development and evaluation of future hearing device technology
When: Thursday December 12th at 4PM <<< Note special time and date due to travel constraints
Where: CCRMA Seminar Room
Why: We need better tools to research our hearing, and how to improve it.
Bring your favorite ears to CCRMA and we'll talk about how to make them better.
---
Open tools for the development and evaluation of future hearing device technology
Hendrik Kayser (Oldenburg)
I will talk about efforts carried out in different projects (funded by the German Research Foundation and the NIH) running at the University of Oldenburg in collaboration with several partners aiming to provide the hearing research community with open tools for the investigation and development of methods and algorithms for future hearing devices and their evaluation in realistic environments.
The open Master Hearing Aid (openMHA, http://www.openmha.org/) software platform enables real-time, low-latency audio signal processing on a variety of computer systems running Linux, Windows and macOS operating systems. It contains a versatile set of basic and advanced methods for the creation of signal processing chains for hearing devices as well as tools and documentation to design own setups on different user levels, including novel algorithms and fitting procedures. Portable setups are of particular interest for use in field studies that allow insight into hearing aid users’ requirements and the investigation of new methods in real-word scenarios. For this, a portable hearing aid research platform based on openMHA is being developed, which is freely configurable and enables control from external devices such as a smartphone.
To test novel algorithms and devices in the lab, novel evaluation methods are required that are more representative of real-life hearing device use than the established lab-based methods, while preserving full control over experimental parameters. For this, interactive virtual audio-visual environments and evaluation methods are being developed, which are based on a self-developed open software toolbox for acoustic scene creation and rendering (TASCAR, http://tascar.org). The software architecture allows incorporating openMHA as a virtual hearing aid into TASCAR-based virtual environments for algorithm evaluation and interactive subject testing.
In this talk I will introduce openMHA as well as TASCAR-based virtual environments, and give relevant application examples.
----------------------------
Hendrik Kayser is a postdoctoral researcher in the "Digital Hearing Devices" group lead by Prof. Volker Hohmann at the Department of Medical Physics and Acoustics, University of Oldenburg. During his PhD thesis in Physics he combined principles of auditory processing and machine learning for sound source localization. His further experience and ongoing projects are in the fields of low-latency real-time signal processing for hearing devices, spatial signal enhancement and acoustic parameter estimation, sound source separation, and computational scene analysis.