The below Carl Zeiss Foundation-funded five year research project at Ilmenau Technical University (July 2023 - June 2028 resp. October 2023 - September 2028, depending on which website you consult) got my Akida-detector sensor tingling, but of course it could be a false positive, given my error-prone sensor doesn’t yet have ‘Akida inside’…
Also involved in the project are another three institutions from the German state of Thüringen (Thuringia): Friedrich Schiller University of Jena, the Fraunhofer Institute for Digital Media Technology IDMT and the IMMS Institute for Microelectronic and Mechatronic Systems.
On a side note: did the foundation’s budget not allow for another “measly” € 1000 to round it up to € 5 million instead of funding merely “4.999.000 €?”
(In German we use decimal points instead of commas to structure large numbers and in turn use a comma instead of a decimal point, just in case you were wondering…)
Well, who knows, the Carl Zeiss Foundation may have wisely decided to invest the money saved in BRN shares instead - by the time Tradegate’s closing bell rang on Friday, that amount would have bought them 4344 shares, which will be worth so much more one day.
Only 41% of people with hearing loss use a hearing aid. NeuroSensEar aims to improve the acceptance of and provision with hearing aids by increasing performance. To this end, bio-inspired sensor technology is combined with sound analysis.
www.carl-zeiss-stiftung.de
Neumorphic acoustic sensor technology for high-performance hearing aids of tomorrow (NeuroSensEar)
Only 41% of people with hearing loss use a hearing aid. NeuroSensEar aims to improve the acceptance of and provision with hearing aids by increasing performance. To this end, bio-inspired sensor technology is combined with sound analysis.
Goals
More than 11% of people in the EU are affected by hearing loss, but only 41% use a hearing aid. The reasons for this are problems in understanding speech and in adapting the devices to the patient. The project goal is therefore to improve the acceptance of and provision of hearing aids by significantly increasing their performance. The adaptation to the patient and different hearing situations is to be greatly facilitated and automated. To this end, principles of biological information processing will be integrated into hearing aid technology and auditory perception will be investigated for better hearing comprehension. Persons with hearing impairment should thus regain their hearing perception to a large extent.
Bio-inspired sensory technology and speech processing are mutually linked for this purpose. The bio-inspired system, i.e. neuromorphic system, will consist of MEMS-based sensor technology and artificial neural networks. The interdisciplinary project team is concerned with sensor development, bio-inspired control and adaptation of the sensor technology, speech processing, situation recognition for sensor control as well as lifelong learning for it and its use in a new hearing aid technology.
Involved persons:
Dr. Phil-Alan Gärtig
Program Manager
Phone: +49 (0)711 - 162213 - 10
E-mail:
phil-alan.gaertig@carl-zeiss-stiftung.de
Prof. Dr. Martin Ziegler
Technische Universität Ilmenau
Detailed information:
Target group: | Professors |
---|
Funding budget: | 4.999.000 € |
---|
Period of time: | Juli 2023 - Juni 2028 |
---|
Funded institution:
TU Ilmenau/ari Investigation of the smart sensor compared to a classic microphone In the large-scale collaborative research project NeuroSensEar, the
www.tu-ilmenau.de
09.05.2023
NeuroSensEar: Bio-inspired technology for highly efficient hearing aids
TU Ilmenau/ariInvestigation of the smart sensor compared to a classic microphone
In the large-scale collaborative research project NeuroSensEar, the Technische Universität Ilmenau is developing a biology-inspired technology that makes highly efficient hearing aids possible. People with hearing impairment should thus largely regain their ability to perceive hearing. The Carl Zeiss Foundation is funding the five-year research project, which starts in October and has a total volume of more than 5.5 million euros, with almost five million euros as part of the "CZS Breakthroughs" funding program. In addition to TU Ilmenau, the Friedrich Schiller University of Jena, the Fraunhofer Institute for Digital Media Technology IDMT and the IMMS Institute for Microelectronic and Mechatronic Systems are involved in the project.
More than eleven percent of people in the European Union are affected by hearing loss, but only 41 percent of them use a hearing aid because they cannot understand speech satisfactorily even with a device. Others forgo a hearing aid because fitting by a professional is often costly and time-consuming. Conventional hearing aids analyze sound and emphasize important signals, especially speech, over unimportant signals and background noise, such as street noise or background sounds. To do this, all sound signals are filtered and analyzed. But hearing aids are reaching their limits.
Challenge for current hearing aids: filtering complex sound situations
The more sound sources there are, for example in train station halls or restaurants, the more difficult it is for hearing aids to separate the important signal speech from background noise. In addition,
the programs that improve hearing perception even in such complex listening scenarios consume so much computing power and energy that they cannot be used in battery-powered hearing aids.
Goal: Increase acceptance of hearing aids through intelligent, adaptive sensor technology
The goal of
the NeuroSensEar project ("Neuromorphic acoustic sensor technology for powerful hearing aids of tomorrow") is to increase the acceptance of hearing aids among the general public and thus improve care for people with hearing loss.
At the heart of the innovative technology is an intelligent sensor that automatically adapts to the respective hearing situation and the individual hearing loss of the user. For this purpose, the sensor is equipped with a control system that reacts in real time to changing acoustic scenarios, for example in the living room or at the train station, and learns new hearing scenarios throughout its life. Its lifelong learning ability enables the sensor to pick up mainly the important signals more and more effectively: Speech signals, for example, would be picked up with higher gain and sensitivity, while background noise would be attenuated. By enabling better listening comprehension, future hearing systems would thus not only be more powerful: More computing power with the same, if not lower, energy consumption also makes them more efficient.
The human ear as a model
TU Ilmenau/ari
The smart sensor will be at the heart of innovative hearing aids
The innovative sensor can be used in hearing aids of all kinds: Mainly in hearing aids and hearables, which are headphones with additional functions such as speech comprehension enhancement and noise suppression, but also in cochlear implants, i.e. hearing prostheses for the severely hearing impaired for whom hearing aids are not sufficient, and for the deaf whose auditory nerve is still functional. In developing the innovative hearing aid technology, the scientists were inspired by the biological information processing of the human auditory system. In particular, the strong adaptability of the sensor technology, the signal processing integrated into the sensor and the close link between sensor and sound processing borrow from the capabilities of the human auditory system.
Strong Thuringian research network
Three groups of the TU Ilmenau, the Friedrich Schiller University of Jena, the Fraunhofer Institute for Digital Media Technology IDMT and theIMMS Institut für Mikroelektronik- und Mechatronik-Systeme gemeinnützige GmbH are working together together. They are developing the smart sensor, researching bio-inspired signal processing, analyzing auditory perception and working on increasing the efficiency of the sensor system and on lifelong learning. The Animal Physiology group at Friedrich Schiller University Jena is studying the biological principles of auditory signal processing, among other things, using various insects.
About the Carl Zeiss Foundation:
The Carl Zeiss Foundation aims to create scope for scientific breakthroughs. As a partner of excellent science, it supports both basic research and application-oriented research and teaching in the STEM disciplines (mathematics, computer science, natural sciences and technology). Founded in 1889 by the physicist and mathematician Ernst Abbe, the Carl Zeiss Foundation is one of the oldest and largest private science-funding foundations in Germany. It is the sole owner of Carl Zeiss AG and SCHOTT AG. Its projects are financed from the dividend distributions of the two foundation companies.
Contact
Dr. Claudia Lenk
Micro- and Nanoelectronic Systems
+49 3677 69-1589
claudia.lenk@tu-ilmenau.de
www.neurosensear.de
NeuroSensEar Neuromorphic acoustic sensor technology for high-performance hearing aids of tomorrow More than 11% of people in the EU are affected by
www.tu-ilmenau.de
NeuroSensEar
Neuromorphic acoustic sensor technology for high-performance hearing aids of tomorrow
More than 11% of people in the EU are affected by hearing loss, but only 41% use a hearing aid due to continued problems with speech understanding and fitting the devices. The NeuroSensEar proket aims to improve the acceptance of and provision of hearing aids by significantly increasing their performance and greatly facilitating and automating their adaptation to the patient and different listening situations.
To achieve this,
principles of biological information processing are integrated into hearing aid technology and interactive outputs for better listening comprehension are investigated, so that persons with hearing impairment largely regain their ability to perceive hearing.
Project partner
Technische Universität Ilmenau
Friedrich Schiller University Jena
Fraunhofer Institute for Digital Media Technology IDMT
IMMS Institut für Mikroelektronik- und Mechatronik-Systeme gemeinnützige GmbH (IMMS GmbH)
Goal and implementation
The aim is to significantly improve the acceptance and supply of hearing aids by significantly increasing their performance and greatly facilitating and automating their adaptation to the patient and different hearing situations. This will, in the long run, help to reduce the economic costs and the severe social consequences in terms of sustainable and efficient health care.
As a breakthrough, we aim to solve two main problems of current hearing aids:
1. hearing comprehension in difficult listening situations with many sound sources and low signal-to-noise ratios.
and
2. the lifelong ability to recognize, learn and act in new listening situations and demands, according to a continuous adaptation to the wearer and his/her hearing and the changing life/environment surrounding him/her.
Scientific advisory board
Prof. Dr. Eckehard Schöll (TU Berlin, BCCN, PIK):
Expert for nonlinear dynamics
Prof. Dr. Herbert Jäger (U. Groningen):
Expert in reservoir computing/ML
Dr. Bipin Rajendran (Kings College London):
Expert for spike-based neural networks (for sensor technology)
Prof. Dr. Elisabetta Chicca (U. Groningen):
Expert in neuromorphic VLSI electronics
Prof. Dr. Manfred Kössl (Uni Frankfurt):
Expert for signal processing through the cochlea
Prof. Dr. Christian Dobel (Uniklinikum Jena):
ENT physician and expert in auditory perception
Transfer Advisory Board
Audifon (Dr. Tamas Harczos): Thuringian hearing aid manufacturer (KIND Group)
XFAB (Dr. Gabriel Kittler): Manufacturer for MEMS and semiconductor integrated circuits
Contact
Prof. Martin Ziegler FG Micro- and nanoelectronic systems
Dr. Claudia Lenk FG Micro- and Nanoelectronic Systems Head of Biosensorics Working Group Project Coordination
claudia.lenk@tu-ilmenau.de
+49 3677 69 1589
Funding information
Funding source: Carl Zeiss Foundation under the funding line "CZS Breakthroughs: Life Science Technologies - Sensor Technology".
Funding volume: 4.99 million euros
Funding period: 01.10.2023 - 30.09.2028
Some observations from my side:
-
One of the members of the scientific advisory board happens to be
Dr. Bipin Rajendran (Kings College London):
Expert for spike-based neural networks (for sensor technology)
On reading this, I immediately recalled the link to a King’s College press release titled
King's engineers to explore brain-inspired computing in 6G telecommunications I posted a couple of weeks ago:
https://thestockexchange.com.au/threads/brn-discussion-ongoing.1/post-329286
In case you wonder whether the researchers there may not even have heard about Brainchip, rest assured they have:
@Fact Finder had previously dug up the fact that the King’s College researchers have known about Akida since 2019.
https://thestockexchange.com.au/threads/brn-discussion-ongoing.1/post-215109
By the way, King’s College London is currently looking for both a two year post-doc position to work with Dr Bipin Rajendran on neuromorphic system design for wireless communications a well as another one with Professor Osvaldo Simeone on neuromorphic computing for wireless communications.
- Dr. Tamas Harczos from the
Thuringian hearing aid manufacturer Audifon (KIND Group) is listed as one of two member of the transfer advisory board.
KIND (www.kind.com) is HUGE! Most adults in German-speaking countries will be familiar with their slogan “Ich hab ein KIND im Ohr”.
“KIND is the market leader in Germany and one of the world's leading companies in hearing aid acoustics. With the brands KIND and audifon it covers the entire value chain of hearing aid acoustics: from development and production to qualified fitting.”
- If Akida is indeed the reason behind the innovative hearing technology to be developed with the help of funding by the Carl Zeiss Foundation, it is practically a sure bet IMO that one of Brainchip’s current or prospective (provided the research project will deliver promising interim results)
NDA partners is/will be either
Carl Zeiss AG or
Schott AG (or both).
- Before I am getting all too excited, though, here is a
caveat that I just noticed before hitting the send button when googling the second transfer advisory board’s industry partner:
X-FAB. On their website
www.xfabulous.com, the company states “
As a pure-play foundry, we provide manufacturing and strong design support services to our customers that design analog/mixed-signal integrated circuits (ICs) and other semiconductor devices for use in their own products or the products of their customers” and “
We offer our customers a broad spectrum of process technologies and sufficient manufacturing capacity (100,000 wafer starts per month) to serve the growing demand for analog/mixed-signal applications. Our wafers are manufactured based on advanced modular CMOS and SOI processes with technologies ranging from 1.0 down to 0.13 micrometer. In addition, we offer special-purpose technologies such as SiC and MEMS processes.”
So does that exclude Akida then, being fully digital?
5 min later… Strangely, I just discovered a second web domain by the same company (
www.xfab.com), which has an image that reassured me I may still be on the right track, even though the company’s slogan is “Your specialty foundry FOR THE ANALOG WORLD”, as the foundry also seems to offer to manufacture in fully digital:
This and more info on
https://www.xfab.com/technology (which is way above my pay grade but will be helpful to some of you in deciding whether the X-FAB connection is a KO criterion).