Ghent University

Engineer in real-time signal processing for auditory neural networks (26678)

Unspecified
Baan opslaan

Over de werkgever

Ghent University is one of the top 100 universities in the Dutch language area, with more than 44,000 students and 15,000 staff members.

De pagina van de werkgever bekijken

   →  Apply before 15/12/2023 (DD/MM/YYYY) 23:59 (Brussels Time)
   →  Faculty of Engineering and Architecture  
   →  Department: EA05 - Informatietechnologie
   →  Occupancy rate: 100%
   →  Number of positions: 1    
   →  Type of employment: Contract of limited duration 
   →  Term of assignment: 18 maanden 
   →  Wage scale:  WM1 to WM4 (master degree or doctoral decree) 
   →  Required diploma: Master or PhD 
 

ABOUT GHENT UNIVERSITY

The research groups are part of Ghent University. Ghent University is a world of its own. Employing more than 15,000 people, it is actively involved in education and research, management and administration, as well as technical and social service provision on a daily basis. It is one of the largest, most exciting employers in the area and offers great career opportunities. With its 11 faculties and more than 80 departments offering state-of-the-art study programmes grounded in research in a wide range of academic fields, Ghent University is a logical choice for its staff and students.

The Internet Technology and Data Science lab (Prof. Torfs) and Hearing Technology @WAVES lab (Prof. Verhulst) in the Dept. of information technology explore and innovate at the interface between embedded systems and hearing technologies. This unique collaboration ensures that the latest neural-network-based audio processing can be embedded in the next generation hearables and hearing aids. The research groups have labs for analog and digital electronics, acoustic measurement setups, and GPU clusters for training neural networks. You will have colleagues who work in low-power and digital electronics, auditory neuroscience, and audio signal processing. 
By joining the expertise of both groups, the EarDiTech project emerged. Inside this project we develop EEG-based measuring hardware and diagnostic software to detect the first signs of hearing damage. The outcome of the EEG measurements can be used to infer individualized compensation algorithms that could be implemented in hearing aids and hearables. Optimization for size and real-time implementation such that the algorithms can run on constraint devices is challenging and will be your main task. 
The innovate character and great market potential (e.g., as a future spin-off project) of this approach resulted in a very strong supported by UGent Technology Transfer and the European Innovation Council.
More info: 
Cordis and EarDiTech
IDLab and Hearing Technology Lab


YOUR TASKS

We are looking for an engineer in embedded systems with an interest in audio and a sense of entrepreneurship to join our EarDiTech team. In this innovation project, you will implement next-generation neural-network-based audio signal processing on FPGA and on the hardware of hearing aids and hearables

-    You are responsible for the technical design of the real-time algorithms implementing  convolutional neural networks for audio applications on constraint AI processors.
-    You test your designs both on the target hardware and in software (benchmarking, efficiency, sound quality, memory, and power consumption).
-    You ensure that the trade-off between sound quality, real-time and low-power performance is always considered.
-    You document the software and on-hardware implementation and can communicate your decisions regarding the next development steps to others in the team.
-    You work together with other engineers within the team, but also with subcontractors, the clinical trial lead and technology transfer experts.

WHAT WE ARE LOOKING FOR

-    You hold a Master’s degree or a thesis-based doctorate in Engineering (Information Technology, Computer Science, Electrical Engineering, Acoustics), Physics or Mathematics.
-    You master the topics of signal processing (DSP, filtering) and embedded systems and are interested in non-linear time domain equations.
-    You are strong in programming (Python or Matlab), and previous experience with machine learning techniques (Keras or Pytorch) are necessary and knowledge of low-level programming languages (C/C++) to implement the design on hardware.
-    You have ideas on how technical specifications of software applications can lead to neural network or architecture optimization for constraint devices.
-    You are interested in evaluating the soft- and hardware audio applications you design and can set up the necessary objective or sound perception experiments in a team.
-    Excellent written and oral English skills are required.

WHAT WE CAN OFFER YOU

-    We offer you a contract of definite duration for the period from January 2nd 2024 to July 31st 2025
-    Your contract will start on 2/01/2024 at the earliest.
-    Your remuneration will be determined by salary scale WM1 to WM4 (if you hold a Master’s degree) or salary scale PD1 to PD4 (if you hold a doctorate).  Click here for more information about our salary scales.
-    All Ghent University staff members enjoy a number of benefits, such as a wide range of training and education opportunities, 36 days of holiday leave (on an annual basis for a full-time job) supplemented by annual fixed bridge days, bicycle allowance and eco vouchers. Click here for a complete overview of all the staff benefits (in Dutch).


INTERESTED?

Apply online through the e-recruitment system before the application deadline (see above). We do not accept late applications or applications that are not submitted through the online system.

Your application must include the following documents:

-    In the field ‘CV’: your CV and an overview of your study results (merged into one pdf file)
-    In the field ‘Cover letter’: your application letter in pdf format
-    In the field ‘Diploma’: a transcript of the required degree (if already in your possession). If you have a foreign diploma in a language other than our national languages (Dutch, French or German) or English, please add a translation in one of the mentioned languages.
-    In the field “other documents”: the name and contact details of 2 people willing to provide a reference and an overview of your study results.


Note that the maximum file size for each field is 10 MB.

As Ghent University maintains an equal opportunities and diversity policy, everyone is encouraged to apply for this position.


MORE INFORMATION

For more information about this vacancy, please contact Prof. Verhulst (s.verhulst@UGent.be). Important: do NOT send your application by email, but apply online.

Do you have a question regarding the online application process? Please read our FAQ or contact helpdesk (dpo@ugent.be - tel. 09/264 98 98).

Informatie over de vacature

Functienaam
Engineer in real-time signal processing for auditory neural networks (26678)
Werkgever
Locatie
Sint-Pietersnieuwstraat 33 Gent, België
Gepubliceerd
2023-11-29
Uiterste sollicitatiedatum
Unspecified
Soort functie
Baan opslaan

Meer vacatures bij deze werkgever

Over de werkgever

Ghent University is one of the top 100 universities in the Dutch language area, with more than 44,000 students and 15,000 staff members.

De pagina van de werkgever bekijken

Dit vind je misschien ook interessant

...
Understanding Users to Optimise 3D Experiences Centrum Wiskunde & Informatica (CWI) Leestijd: 5 min
...
Control Systems: The Key to Our Automated Future? Max Planck Institute for Software Systems (MPI-SWS) Leestijd: 5 min
Meer stories