Stefan Thaleiser erwarb 2015 seinen Bachelor- und 2018 seinen Master-Abschluss in Elektrotechnik und Informationstechnik an der Ruhr-Universität Bochum. Seit 2018 ist er am Institut für Kommunikationsakustik der Ruhr-Universität Bochum tätig, wo er derzeit an seiner Promotion arbeitet. Seine Forschungsinteressen umfassen die ein- und mehrkanalige Sprachsignalverarbeitung für Hörgeräte, CI und andere Hörsysteme und basieren auf Methoden der statistischen Signalverarbeitung und des maschinellen Lernens.

Veröffentlichungen

Thaleiser, S., Enzner, G., Martin, R., & Chinaev, A. (2025). Common-Gain Autoencoder Network for Binaural Speech Enhancement. In IEEE Open Journal of Signal Processing, 1–10. https://doi.org/10.1109/OJSP.2025.3633577 [Link to Audio Files] [Link to Code Capsule]

Chinaev, A., Enzner, G., & Thaleiser, S. (2025). Optimization of Feature and Loss Exponents for Lightweight DNN-based Binaural Speech Enhancement. Speech Communication; 16th ITG Conference, Berlin, Germany, 2025, pp. 161-165. https://doi.org/10.30420/456617033

Chinaev, A., Spitz, T., Thaleiser, S., & Enzner, G. (2024). Matrix Study of Feature Compression Types and Instrumental Speech Quality Metrics in Ultra-Light DNN-Based Spectral Speech Enhancement. 2024 18th International Workshop on Acoustic Signal Enhancement (IWAENC), 11–15. https://doi.org/10.1109/IWAENC61483.2024.10694242

Thaleiser, S., & Enzner, G. (2023). Binaural-Projection Multichannel Wiener Filter for Cue-Preserving Binaural Speech Enhancement. IEEE/ACM Transactions on Audio, Speech, and Language Processing, 31, 3730–3745. https://doi.org/10.1109/TASLP.2023.3317569

Thaleiser, S., Chinaev, A., Martin, R., & Enzner, G. (2022). Dual-Compression Neural Network with Optimized Output Weighting For Improved Single-Channel Speech Enhancement. In 2022 International Workshop on Acoustic Signal Processing (IWAENC) (pp. 1-5). https://doi.org/10.1109/IWAENC53105.2022.9914737 | Link zu Audio Files

Thaleiser, S., & Enzner, G. (2022). Binaural Wind-Noise Tracking with Steering Preset. In 30th European Signal Processing Conference (EUSIPCO 2022): Proceedings (pp. 70-74). https://doi.org/10.23919/EUSIPCO55093.2022.9909804

Thaleiser, S., & Enzner, G. (2021). Cue-Preserving MMSE Filter with Bayesian SNR Marginalization for Binaural Speech Enhancement. In ICASSP 2021 - 2021 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP) (pp. 6124–6128). IEEE. https://doi.org/10.1109/ICASSP39728.2021.9414956

Thaleiser, S., & Enzner, G. (2020). A Computationally Light Algorithm for Bayesian Speech Enhancement with SNR Marginalization. In ICASSP 2020 - 2020 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP) (pp. 6209-6213). IEEE. https://doi.org/10.1109/ICASSP40776.2020.9054611