Faculty of Science

Open Opportunities

Expressive Speech Synthesis for Accessibility and Health University of Zurich Michael Krauthammer Developing AI models to create natural-sounding, expressive speech synthesis systems for healthcare applications. This project focuses on improving synthetic voices to capture emotion, tone, and non-verbal cues, enabling effective communication for individuals with speech impairments. Behavioural and Cognitive Sciences, Engineering and Technology Master Thesis, Semester Project

Controllable Music Generation and Editing with AI for Health, Creativity, and Wellbeing University of Zurich Michael Krauthammer This project explores controllable music generation and editing using cutting-edge AI techniques. Instead of generating entire songs with a single text prompt, we aim to create fine-grained, temporally controlled music where specific aspects (e.g., melody, chords, drum patterns, musical style) can be independently specified, edited, and regenerated. Such controllable music systems open exciting applications not only in creative industries but also in healthcare and wellbeing, supporting adaptive music therapy, emotional regulation, and accessible creative tools for individuals with disabilities. Arts, Behavioural and Cognitive Sciences, Engineering and Technology Bachelor Thesis, Master Thesis, Semester Project

Computational Methods for Protein Design and Fitness Optimization University of Zurich Michael Krauthammer We are currently looking for Master’s students with background in machine learning (or related computational field) for a project on Protein Fitness Optimization. Artificial Intelligence and Signal and Image Processing, Computational Structural Biology Master Thesis, Semester Project

Automating the Detection of DNA Replication Forks in TEM Images Using Deep Learning University of Zurich Lopes This project aims to develop a deep learning model to automate the identification of DNA replication intermediates (RIs) in high-resolution Transmission Electron Microscopy (TEM) images—a process currently reliant on manual review. Leveraging a rich dataset from the Lopes lab at the University of Zurich, the model will classify image tiles containing RIs and rank them by prediction confidence to streamline analysis. The project also includes implementing interpretability tools to uncover features associated with RIs. It is ideal for candidates with strong computational skills, experience in deep learning (e.g., PyTorch or TensorFlow), and an interest in interdisciplinary research at the interface of biology and AI. Electronmicroscopy, Genome Structure, Image Processing, Neural Networks, Genetic Alogrithms and Fuzzy Logic, Protein Targeting and Signal Transduction, Systems Biology and Networks Internship, Master Thesis

Mapping spruce density using aerial imagery and deep learning ETH Zurich Forest Resources Management Other organizations: EcoVision Lab Climate change is increasing tree mortality due to drought and biotic infestations, but current detection methods are limited by data availability and low transferability. This study aims to use deep learning with true color near-infrared RGBI aerial imagery to detect spruce mortality in mixed forests. By integrating field inventories and RGB imagery, the method will be analyzed using R or ArcGIS Pro to accurately assess vegetation conditions. Environmental Sciences, Geomatic Engineering, Information, Computing and Communication Sciences Master Thesis, Semester Project

Enhancing tree species identification using multi-view Convolutional Neural Networks ETH Zurich Forest Resources Management Other organizations: EcoVision Lab Tree species identification is crucial for biodiversity monitoring, forest management, and understanding ecological processes. Advances in computer vision and deep learning have enabled the use of multi-view convolutional neural networks (CNNs) to classify species by integrating complementary information from different views. This thesis explores the integration of multi-view data and citizen science images to develop a scalable, high-accuracy tree species identification framework. By addressing challenges related to data variability and leveraging diverse georeferenced plant images, the study aims to enhance the training and generalization of multi-view CNN models. Computer Vision, Forestry Sciences, Photogrammetry and Remote Sensing Master Thesis

Master Thesis: Fraternal Inhibition in Songbird Brother ETH Zurich Institute of Neuroinformatics Our research group aims to enhance the understanding of human language acquisition and development using songbird as model. We are particularly interested in the evolutionary aspects of language, where two developmental tendencies are observed: convergent and divergent evolution. Convergent evolution refers to the simplification of language complexity, similar to how infants gradually acquire human language. Conversely, divergent evolution involves an increase in complexity, akin to teenagers creating and using novel words to establish unique identities. We propose to investigate whether similar effects are observable in animal vocalization learning, specifically in song learning of zebra finches and to explore the effect of social interaction. To facilitate this investigation, our team has developed a "birdpark," a multimodal recording system that provides a naturalistic social environment for observing and recording multiple zebra finches within a dynamic group context. Learning, Memory, Cognition and Language, Linguistic Processes (incl. Speech Production and Comprehension), Sensory Systems, Signal Processing, Zoology ETH Zurich (ETHZ), Master Thesis, Semester Project

Developing a real-time state-machine for closed-loop brain-machine interfaces ETH Zurich Neurotechnology Developing a state-machine Simulink model to be deployed at MathWorks SpeedGoat real-time target machine for closed-loop brain-machine interface (BMI). The state-machine will control the closed-loop BMI peripherals and synchronise the data flow. Peripherals include neural recorders & stimulators, data analysis cluster, video cameras and experimental chamber. Experimental chamber (variety of servos, steppers, sensors etc.) will be controlled with built-in FPGA and GPIO of SpeedGoat machine. Other peripherals are connected with serial bus. Acquired data needs to be organized and stored in datasink unit. Skills: Matlab Simulink, state-machines, FPGA programming, serial communication protocols, data synchronisation Please send your CV and transcript along with your application. Arithmetic and Logic Structures, Computer Communications Networks, Digital Systems, Electrical Engineering, Input, Output and Data Devices, Logic Design Bachelor Thesis, Internship, Master Thesis, Semester Project

Developing a GUI for cross-modality brain-machine interfaces ETH Zurich Neurotechnology Programming a graphical user interface (e.g. in Qt/C++) which can handle and process the data acquired in our brain-machine interface (BMI) experiments. The data includes high-density brain activity recordings from hundreds of recording channels, neural-stimulation events, 3D&4D data coming from MRI scans of the subject implanted with BMI. The backend will be programmed in Python where you also need to connect supporting tools (e.g. Blender) via Python. Please send an email with your CV and transcript of records attached. Electrical and Electronic Engineering, Software Engineering Bachelor Thesis, ETH Zurich (ETHZ), Internship, Master Thesis, Semester Project