Personal

Welcome to the homepage of Wilfried Michel.

Currently I am PhD student within the Human Language Technology and Pattern Recognition Group at the Chair of Computer Science 6 of the RWTH Aachen University.

The main research topics of the chair are

Pattern Recognition
Automatic Speech Recognition
Statistical Machine Translation
Image Processing

I am part of the speech recognition group. The focus of my work is discriminative training for acoustic modelling.

You can find me in room 6115 of our department, call me at +49 241 80 21636 or write an e-mail to <surname>@i6.informatik.rwth-aachen.de

Short CV

Apr 2019 - today	Speech Scientist at AppTek GmbH, Aachen
Mar 2016 - Dec 2021	Research Assistant and PhD Student at the Lehrstuhl für Informatik 6, RWTH Aachen University
Apr 2013 - Feb 2016	Research Assistant and B.Sc./M.Sc. Student at the Computational Condensed Matter Physics Group, University of Cologne
Oct 2012 - Aug 2014	Teaching Assistant for the Beginners Lab Cours ("Praktikum A"), University of Cologne
Oct 2010 - Nov 2015	Undergrad Studies in Physics at University of Cologne

A long version of my CV is available upon request.

Teaching

Seminar Supervision

Proseminar Datenkompression, WS16/17 - SS21, Topics: "Audio Compression".
Selected Topics in HLTPR, SS17 - SS21, Topics: "Sequence Discriminative Training", "Speaker Diarization".

Tutorial Classes

Computerphysik, SS15.
Theoretische Physik in zwei Semestern II, WS14/15.

Lab Courses

ASR Software Lab, WS18/19 - WS20/21.
Praktikum A, WS12/13 - SS14, Topics: "Electricity".

Publications

2023

D. Mann, T. Raissi, W. Michel, R. Schlüter, and H. Ney. End-to-End Training of a Neural HMM with Label and Transition Probabilities. In IEEE Automatic Speech Recognition and Understanding Workshop (ASRU), October 2023. Preprint Arxiv:2310.02724.

2022

F. Meyer, W. Michel, M. Zeineldeen, R. Schlüter, and H. Ney. Automatic Learning of Subword Dependent Model Scales. In Interspeech, Incheon, Korea, September 2022.
W. Zhou, W. Michel, R. Schlüter, and H. Ney. Efficient Training of Neural Transducer for Speech Recognition. In Interspeech, pages 2058-2062, Incheon, Korea, September 2022. [poster].
N. Wynands, W. Michel, J. Rosendahl, R. Schlüter, and H. Ney. Efficient Sequence Training of Attention Models using Approximative Recombination. In IEEE International Conference on Acoustics, Speech, and Signal Processing (ICASSP), Singapore, May 2022. Preprint arXiv:2110.09245.
M. Zeineldeen, J. Xu, C. Lüscher, W. Michel, A. Gerstenberger, R. Schlüter, and H. Ney. Conformer-based Hybrid ASR System for Switchboard Dataset. In IEEE International Conference on Acoustics, Speech, and Signal Processing (ICASSP), Singapore, May 2022. [poster].

2021

P. Vieting, C. Lüscher, W. Michel, R. Schlüter, and H. Ney. On Architectures and Training for Raw Waveform Feature Extraction in ASR. In IEEE Automatic Speech Recognition and Understanding Workshop (ASRU), pages 267-274, Cartagena, Colombia, December 2021.
A. Zeyer, A. Merboldt, W. Michel, R. Schlüter, and H. Ney. Librispeech Transducer Model with Internal Language Model Prior Correction. In Interspeech, pages 2052-2056, August 2021. Arxiv:2104.03006.
M. Zeineldeen, A. Glushko, W. Michel, A. Zeyer, R. Schlüter, and H. Ney. Investigating Methods to Improve Language Model Integration for Attention-based Encoder-Decoder ASR Models. In Interspeech, pages 2856-2860, August 2021. [slides] [video].

2020

W. Michel, R. Schlüter, and H. Ney. Early Stage LM Integration Using Local and Global Log-Linear Combination. In Interspeech, pages 3605-3609, Shanghai, China, October 2020.
W. Zhou, W. Michel, K. Irie, M. Kitza, R. Schlüter, and H. Ney. The RWTH ASR system for TED-LIUM release 2: Improving Hybrid HMM with SpecAugment. In IEEE International Conference on Acoustics, Speech, and Signal Processing (ICASSP), pages 7839-7843, Barcelona, Spain, May 2020.
W. Michel, R. Schlüter, and H. Ney. Frame-Level MMI as a Sequence Discriminative Training Criterion for LVCSR. In IEEE International Conference on Acoustics, Speech, and Signal Processing (ICASSP), pages 6904-6908, Barcelona, Spain, May 2020.

2019

W. Michel, R. Schlüter, and H. Ney. Comparison of Lattice-Free and Lattice-Based Sequence Discriminative Training Criteria for LVCSR. In Interspeech, pages 1601-1605, Graz, Austria, September 2019.
C. Lüscher, E. Beck, K. Irie, M. Kitza, W. Michel, A. Zeyer, R. Schlüter, and H. Ney. RWTH ASR Systems for LibriSpeech: Hybrid vs Attention. In Interspeech, pages 231-235, Graz, Austria, September 2019.

2018

M. Kitza, W. Michel, C. Boeddeker, J. Heitkaemper, T. Menne, R. Schlüter, H. Ney, J. Schmalenstroeer, L. Drude, J. Heymann, and R. Haeb-Umbach. The RWTH/UPB System Combination for the CHiME 2018 Workshop. In The 5th International Workshop on Speech Processing in Everyday Environments (CHiME-5), pages 53-57, Hyderabat, India, September 2018.

2017

Z. Tüske, W. Michel, R. Schlüter, and H. Ney. Parallel Neural Network Features for Improved Tandem Acoustic Modeling. In Interspeech, pages 1651-1655, Stockholm, Sweden, August 2017.

2016

W. Michel, Z. Tüske, M. A. B. Shaik, R. Schlüter, and H. Ney. The RWTH Aachen LVCSR system for IWSLT-2016 German Skype conversation recognition task. In International Workshop on Spoken Language Translation (IWSLT), Seattle, USA, December 2016.

Full list of publications of the chair.