Lars Thieling, M. Sc.

Scientist

Room: 203
+49 241 80-26959
+49 241 80-22254
thieling(at)iks.rwth-aachen.de

Research

  • Machine Learning Assisted Audio Processing
  • Phase-Aware Speech Enhancement
  • DNN-Based Speech Synthesis
  • Perceptually-Motivated Loss Functions
  • Reconstruction Of Lost Speech Parts (Speech Inpainting)

Teaching

Student's Theses

Publications

Thieling, L. and Jax, P.: Beyond Clean Phase: Using Silence-Generating Phase for DNN-Based Speech Enhancement, in Proceedings of European Signal Processing Conference (EUSIPCO), pp. 111-115, EURASIP, Sep. 2023, ISBN: 978-9-46459-360-0, ISSN: 2076-1465, 10.23919/EUSIPCO58844.2023.10289814


Thieling, L., Nippert, L. and Jax, P.: Using Perceptual Evaluation of Speech Quality (PESQ) Loss for DNN-Based Speech Enhancement, in ITG-Fachtagung Sprachkommunikation, pp. 61-65, VDE, Sep. 2023, ISBN: 978-3-80076-164-7, 10.30420/456164011


Thieling, L. and Jax, P.: Two-Stage Speech Enhancement Using Gated Convolutions, in Proceedings of International Workshop on Acoustic Signal Enhancement (IWAENC), pp. 1-5, IEEE, Sep. 2022, ISBN: 978-1-66546-867-1, 10.1109/IWAENC53105.2022.9914693


Thieling, L. and Jax, P.: Generally Applicable Deep Speech Inpainting Using the Example of Bandwidth Extension, in Proceedings of European Signal Processing Conference (EUSIPCO), pp. 451-455, Aug. 2021, ISBN: 978-9-08279-706-0, ISSN: 2076-1465, 10.23919/EUSIPCO54536.2021.9616099


Thieling, L., Wilhelm, D. and Jax, P.: Recurrent Phase Reconstruction Using Estimated Phase Derivatives From Deep Neural Networks, in Proceedings of IEEE International Conference on Acoustics, Speech, and Signal Processing (ICASSP), pp. 7088-7092, IEEE, Jun. 2021, ISBN: 978-1-72817-605-5, ISSN: 2379-190X, 10.1109/ICASSP39728.2021.9413722


Schäfer, M., Thieling, L. and Vollmer, L.: Metrics for the Evaluation of Audio Quality, in Proceedings of German Annual Conference on Acoustics (DAGA), pp. 1390-1393, 2019, ISBN: 978-3-93929-614-0


Liebich, S., Thieling, L., Fabry, J., Jax, P. and Vary, P.: Natural own-voice perception while wearing headphones - Active Occlusion Cancellation with Equalized Hear-Through, in Proceedings of International Workshop on Acoustic Signal Enhancement (IWAENC), IEEE, Sep. 2018