Prof. Dr. Gerhard Rigoll



The research of Prof. Rigoll (b. 1958) deals with all aspects of pattern recognition for multimodal human-machine interaction. Subfields include speech processing, audiovisual information processing, handwriting recognition, gesture and facial expression recognition, face detection and recognition, object tracking and interactive graphical systems. He is the author or co-author of over 400 publications and has sat on several programming committees. He has been involved in numerous expert panels in Germany and internationally.

After studying technical cybernetics in Stuttgart, he did research at that university’s Fraunhofer Institute. He did his doctorate there in 1986 with a thesis on automatic speech recognition. After that, he was a postdoctoral fellow at IBM Thomas Watson Research Center in Yorktown Heights/USA until 1988. After qualifying as a lecturer in Stuttgart from 1991 to 1993, he was a visiting scientist at the NTT Human Interface Laboratory in Tokyo. From 1993 to 2001, he was professor of computer engineering at Gerhard Mercator University in Duisburg prior to accepting his current position at TUM in 2002.


  • DAGM-Preis der Deutschen Arbeitsgemeinschaft für Mustererkennung (2000)
  • IEEE Senior Membership; Acoustics, Speech, and Signal Processing Society (1998)
  • Heisenberg-Stipendium der DFG (1993)
  • FpF-Preis für die beste Dissertation der Stuttgarter Fraunhofer-Institute (1987)


1. Schenk J, Rigoll G: Mensch-Maschine-Kommunikation, Grundlagen von sprach- und bildbasierten Benutzerschnittstellen. Berlin: Springer-Verlag, 2010.

2. Rigoll G, Schuller B, Müller R, Ablassmeier M, Reifinger S, Poitschke T: “Speech Communication and Multimodal Interfaces”. In: Advanced Man-Machine Interaction. Editor: Kraiss KF. 2005; 141 – 190.

3. Rigoll G: “Combination of Hidden Markov Models and Neural Networks for Hybrid Statistical Pattern Recognition”. In: Hybrid Methods in Pattern Recognition. Editor: Kandel, Bunke. 2001; 113 – 143.

4. Eickeler S, Müller S, Rigoll G: “Recognition of JPEG Compressed Face Images Based on Statistical Methods”. Image and Vision Computing Journal, Special Issue on Facial Image Analysis. 2000; 18(4): 279-287.

5. Rigoll G: “Maximum Mutual Information Neural Networks for Hybrid Connectionist-HMM Speech Recognition Systems”. IEEE Transactions on Speech and Audio Processing, Special Issue on Neural Networks for Speech Processing. 1994; 2(1): 175-184.


© 2022 Technische Universität München Asia
German Institute of Science & Technology - TUM Asia Pte Ltd
CPE Reg. No. 200105229R | Registration Period 13.06.2017 - 12.06.2023

tum arrow up