Index
Development of a deep learning-based phoneme recognizer for noisy speech
For speech intelligibility, consonants have a fundamental importance. Thus, the assumption can be made that automatic phoneme and especially consonant recognition correlates well with human speech intelligibility.
In noisy environments however, speech and especially consonants may be degraded.
In this project, we want to study the effect of noise on speech intelligibility. Therefore, we train a neural network to recognize phoneme based on the TIMIT dataset. We will add diffent noise types and noise levels to the speech signal and study the effect on the recognition rate.
This project requires no preliminary knowledge in deep learning, although may be beneficial. Basic signal processing concepts like sampling theorem and FFT should be present.
Adapting Pyro-NN with SPECT operators
Goal of this project is the implementation of the SPECT Forward and backward projection model in the Pyro-NN Framework. This would enable to include the SPECT reconstruction process into a Neural Network architecture.
Advancing the digital twin method
The aim of this research project was to develop a program that registers an XCAT phantom to a CT scan with a rigid and a non-rigid registration. The registered XCAT Phantom can be used to perform SPECT experiments and simulations without burdening the patient with an additional SPECT examination. To perform the registration the open source software Plastimatch version 1.8.0 was used. The results of the registration were evaluated visually and empirically. The registration was successful in most of the cases, but there were some cases where the rigid registration direction failed.