Research
Current Projects
SPEECHES: Speech Processing in Health Sciences (since 2026/08)
The SPEECHES project aims to develop an on-device speech analysis system for stuttering therapy, enabling continuous monitoring and feedback while ensuring GDPR compliance through local data processing.
Spectrum Sensing (since 2024/04)
This research project focuses on the development of an automatic classification system for radio transmissions based on software defined radio (SDR) and machine learning.
Detection of Wood Rot using Machine Learning (since 2023/11)
The aim of this project is to reduce the workload in sawmills by automatically detecting wood rot on the cross-section of wooden logs using computer vision models.
Semmeldetector (since 2022/06)
The aim of the Semmeldetector is to increase resource efficiency in commercial bakeries by automatically tracking unsold products using computer vision models.
Dementia Detection (since 2021/07)
This project has two main goals: the investigation of (deep) speech markers in machine learning for the assessment of dementia and the automatic speech-based evaluation and digitization of cognitive tests.
Voice Transformation (since 2021/06)
The aim of this project is to improve the intelligibility and naturalness of the voices of laryngectomy patients based on whispered speech by leveraging the advances made in the field of generative models.
NNFuze (since 2021/06)
This project focuses on memory and computational optimization of neural networks for use on embedded systems and addresses the question of how different neural networks can be replaced by or merged to create a single network.
Spirio Sessions (since 2020/06)
Can artificial intelligence collaborate with humans in creative processes at the same level? What would this collaboration look like? These complex questions form the basis for the research conducted by academics at Nuremberg Tech and Nuremberg University of Music as part of an interdisciplinary project funded by LEONARDO - Center for Creativity and Innovation.
Finished Projects
bess-ki.md (2023/11 — 6/2025)
The aim of this project is to support the BESS, a language assessment test for pre-school children, with methods of automatic speech and language processing.
isaki.md (2021/04 — 2023/12)
This project focuses on anomaly detection in critical infrastructures, such as water supply networks. By integrating various data sources, the system aims to improve situational awareness, enabling timely decision-making and resilience against various external threats. (Image: Adobe Stock / #201540971)
