Speech Processing and Machine Learning


Professor Hermann Ney, Chair of Computer Science 6 at RWTH Aachen


The Lecture

No innovation has changed human life more than the development of language and the invention of writing. Spoken and written language form the most important foundation for both human communication and the transfer and storage of knowledge and information, particularly in the globally networked and multilingual digital world of today.

Speech and its processing are considered typical human abilities. The development of automatic systems for speech processing is viewed as a core task in so-called artificial intelligence with the rise of the computer. Typical tasks of speech processing include the recognition and comprehension of spoken language and translation between different languages. The most successful concepts in automatic speech processing until now assume that the computer learns using sample data and works with plausibility evaluations instead of prescribed categorical rules. This occurs with processes from statistical decision theory and machine learning, which also includes artificial neural networks.

Using historical developments, the lecture presented different approaches to automatic speech processing and the current state of science

The Speaker

  • Born in Saarlouis, Saarland, 1952
  • Diplom in physics at the Göttingen University, 1977
  • Doctoral degree in electrical engineering at the Technical University of Braunschweig, 1982
  • Employee at the Philips research laboratories in Hamburg and Aachen, 1977 to 1993
  • Chair of Computer Science 6 at RWTH Aachen, as of 1993
  • Research stays at Bell Labs Murray Hill in 1988/89, UP Valencia in 1996, 2002, and 2007/8, ICSI Berkeley in 1997, Microsoft Research Redmond, 2007
  • Research professorship - DIGITEO chair - at LIMSI-CNRS Paris, 2010 to 2014
  • Participation in various joint projects on speech recognition and speech translation with European (TCStar, QUAERO, EU-BRIDGE, and others) and American funding (GALE, BOLT, BABEL, plus others); more than 700 publicatinos at international conferences and in journals
  • Technical Achievement Award of the IEEE Signal Processing Society in 2005, Award of Honour of the International Association of Machine Translation in 2013, IEEE Fellow in 2009, and Distinguished Lecturer in 2015/6; Int. Speech Communication Association Fellow in 2010, and Distinguished Lecturer in 2012/13