Community Research and Development Information Service - CORDIS

Final Report Summary - ICANHEAR (Improved Communication through Applied Hearing Research)

Improved Communication through Applied Hearing Research (ICanHear, http://www.icanhear.eu)

Coordinator: Prof. Dr.-Ing. Rainer Martin, Institute of Communication Acoustics, Ruhr-Universität Bochum,
44780 Bochum, Germany. Email: rainer.martin@rub.de

ICanHear is a European Union funded Marie Curie Initial Training Network in the field of hearing research. Five academic and two industrial partners in Europe collaborated in this interdisciplinary network to bridge the gap between traditional research areas and to seek novel solutions the field of hearing research and signal processing for hearing instruments. Today many hearing-impaired people have difficulties with speech communication in adverse acoustic conditions. Research in ICanHear focused on paving the way toward the next generation of hearing aids and cochlear implants with improved user experience in complex acoustic environments.

14 doctoral and five postdoctoral research fellows with a wide variety of backgrounds, including audiology, biomedical engineering, information technology and physics were recruited. They were embedded in the functional, supportive and inspiring network of ICanHear, offering excellent opportunities for research and training in hearing science, digital signal processing, clinical audiology and hearing instrument technology. Through the tight integration of academic and industrial partners, the research fellows were familiarised with various stages of product development and were provided with motivation and skills to move from academic research to product development. To this end ICanHear implemented a comprehensive research and training programme including scientific and practical courses, academic and industrial secondments, and public outreach activities.

Research in ICanHear was structured around three themes: Modelling, Evaluation and Application, briefly summarized as follows:

In hearing research, a major challenge is to understand which aspects of auditory processing are impaired beyond those that cause decreased audibility, how individual hearing impairment can be compensated for and, in particular, how speech communication can be improved in natural listening scenarios. Here, computer-based processing models often help to generate hypotheses that can be quantitatively tested for complex systems. In this respect, ICanHear research fellows contributed with their projects in several ways: They focused on computational models of neural responses for speech enhancement, using different concepts of auditory inspired preprocessing; they considered a bio-inspired coding strategy for cochlear implants and developed a computational model of speech intelligibility in hearing-impaired listeners; they finally focused on model-based signal processing for bilateral hearing instruments.

The Evaluation Theme focused on multidisciplinary research on the bridge between modeling of neural signal processing in the impaired auditory system and application of new strategies for improved communication. This theme zoomed in on evaluation of hearing instrument processing and aided hearing. The projects in the Evaluation Theme were dedicated to the development of new methodologies to evaluate advanced signal processing strategies for improved speech intelligibility, directional hearing and comfort of listening with hearing instruments. The fellow projects all focused on objective evaluation methods and instrumental measures, as well as physical or electrophysiological measures based on evoked brain potentials. These projects have opened new avenues for evaluation and optimal fitting of the hearing instrument parameters to the auditory needs of hearing impaired individuals.

In the ICanHear project, one important aim was to make research results, and their benefits in the form of new algorithms and methods, available to the hearing impaired as soon as possible. Therefore, the Application Theme focused directly on developing the newly gained fundamental understanding into technologically relevant algorithms and methods. More specifically, in the projects of the Application Theme research fellows have worked on improving temporal coding and spatial discrimination of cochlear implants to make speech more intelligible and music more enjoyable, they have improved speech signal processing algorithms for dynamic environments, so as to consistently bring to the foreground the voice of the speaker of interest and have integrated the newly available sensors within smartphones and other external devices to improve speech and audio quality.

In addition to their individual research projects all fellows contributed to the development of two central components of ICanHear, the Open Development Platform for Signal Processing in Hearing Aids (ODP) and a common framework for the real-life evaluation of digital signal processing algorithms (Common Evaluation Schemes (CES)). These components have enabled research fellows to develop real-time implementations of signal processing algorithms for audio signals and to evaluate their algorithms with respect to their user benefits. Also, the ODP made it possible to share algorithms and building blocks across the network and it has established a common communication platform between researchers and industry.

In order to deal properly with the ethical aspects of their research projects all ICanHear fellows received training on ethical guidelines and procedures. All relevant project activities were submitted to a mandatory external ethics review at the partners institutions on a case-by-case basis. The preparation and submission of documents for this external review were part of the local fellow training.

All research projects have made significant contributions to the objectives of this programme. A number of innovative approaches, models, and systems were developed, prototyped, and evaluated. Overall, the work has been successful in the development of new, application-driven scientific approaches for cochlear implant coding as well as multi-microphone and multi-device-based speech signal processing for hearing aids, addressing all scientific objectives as planned. This progress has led to new methods that yield better intelligibility and better quality of the enhanced speech and audio signals for users of hearing devices. Through the comprehensive training of ICanHear the research fellows have also made great progress towards their own scientific goals. At the end of ICanHear, all postdoctoral students have acquired a research position at important international research centers.

All research projects resulted in publications in highly ranked and relevant international journals. Furthermore, the results of the activities from the scientific work packages have been presented at various international conferences and workshops. The results have already stimulated various related activities within and outside the ICanHear network.

Finally, the research achievements of the ICanHear project have been showcased at the ICanHear Conference in November 2016. The ICanHear fellows presented their work with short presentations and posters and with demonstrations of real-time implementations of signal-processing algorithms developed in ICanHear on the Speedgoat xPC Target Machine. The conference was open to the scientific community and attracted external specialists from academia and industry in the field of hearing instruments, as well as PhD students and researchers from outside the network. It was an excellent platform for the interaction between ICanHear research fellows and external participants.

More information on ICanHear is available via http://www.icanhear.eu

Contact

Rainer Martin, (Professor)
Tel.: +49 234 3222495
E-mail

Subjects

Life Sciences
Record Number: 199816 / Last updated on: 2017-06-21
Follow us on: RSS Facebook Twitter YouTube Managed by the EU Publications Office Top