Embedded Speech Emotion Recognition on imec neuromorphic processor

Student project at imec (Thesis)

What will you be doing?

In the context of the EU DAIS project (https://dais-project.eu/), we would like to demonstrate EdgeAI privacy-preserving speech emotion recognition to bring the deep neural networks into the intelligent buildings. The main target of this project is to adapt and optimize the algorithms for implementation in the imec neuromorphic processor, SENeCA.  

The main tasks are:

  1. Understanding the target application and algorithm [1] and the target hardware architecture [2].

  2. Optimization of the algorithm for hardware implementation (including but not limited to quantization, sparsification, mapping, etc.).

  3. Embedded hardware implementation.

  4. Power/Latency measurements and benchmarking.

The project can be divided into three phases: 1) Inference only network, 2) Online learning and 3) Federated learning. The student is expected to at least demonstrate the results for phase one of the project. 

[1] Tsouvalas, Vasileios, Tanir Ozcelebi, and Nirvana Meratnia. "Privacy-preserving Speech Emotion Recognition through Semi-Supervised Federated Learning." arXiv preprint arXiv:2202.02611 (2022), https://doi.org/10.48550/arXiv.2202.02611
[2] Amirreza Yousefzadeh, et al. SENeCA: Next-generation Neuromorphic Processor for EdgeAI, TinyML poster, http://dx.doi.org/10.13140/RG.2.2.19651.89124  

What we do for you

Imec is one of the world's leading research institutes in micro and nano-electronics. The Imec-NL lab at Holst Centre is a center of excellence in designing nano-electronics solutions for the Internet of Things and healthcare applications. In this internship project, you will be working on a cutting-edge research project under the supervision of expert researchers from diverse backgrounds. The work of this project will fall under the scope of a European ECSEL project ("DAIS"), and the outputs will be published in deliverables and high-impact journals/conferences (subject to the quality of the work). Imec-NL provides the required equipment, access to lab facilities, a workplace in the Holst Centre at High Tech Campus, and a monthly allowance for running/living expenses during the internship. 

Who you are
  • MSc./Ph.D. students with a relevant background (non-European students are only eligible if they study in the Netherlands).

  • Your are available for 9 months. 

  • You have excellent programming skills in C and Python.

  • Knowledge of deep neural networks and programming in TensorFlow/Keras is a big plus.

  • Good written and verbal English skills.

  • Motivated student, good communicator, easy collaborator, and eager to work independently and expand knowledge in the field.

Interested

Click on 'apply here' to submit your application. You will then be redirected to e-recruiting.