Electrical Engineering and Information Technology

ITG Conference on Speech Communication

September 29th, 2023 | by
Crowd of people in front of poster presentation, listening to the speaker.

© Katharina Uhlig

The fifteenth symposium in a successful series of events on the topics of speech processing, audio and spoken language processing took place from September 20 to 22, 2023 in the SuperC building of RWTH Aachen University.

Central and in the heart of university life, a lively scientific exchange took place with a focus on theory, algorithms, and application relevance of the topic composition. Conference participants exchanged ideas in workshops and benefited from both professional and personal networking. Alternating blocks of lectures and poster presentations, as well as a show & tell session formed elements of the three-day conference.

Audience in line, seated facing presentation from behind.

© Katharina Uhlig

Other highlights included three keynote presentations from renowned individuals with expertise:

For Team Viewer, Dr. Hendrik Witt presented real-life use cases for augmented reality solutions for process optimization in the working world and explained the extent to which wearables and sensors are driving the industrial metaverse. The goal, he said, is digitization far from desks, for example in logistics, manufacturing or field service. Team Viewer sees the metaverse as an interface between the real world, the Internet and technology. Its AR platform Frontline offers augemnted reality, mixed reality and artificial intelligence on smart glasses, smartphones, tablets and other wearables. Empowered so-called frontline workers through this could usher in the next industrial revolution.

Professor Nilesh Madhu of Ghent University vividly explored the evolution of communication technology in his talk; from the use of tedious, hand-made, stochastic models to a plethora of impressive state-of-the-art data-driven approaches. His remarks also allowed a glimpse into the future and opened the discussion on whether the knowledge and experiences of the past are suitable to pave the way there.

Focusing on practical experience and the application of research results, Professor Frank Kurth from the Frauenhofer Institute for Communication, Information Processing and Ergonomics presented concrete safety application scenarios that use various approaches, including Deep Learning, for information extraction. Especially with respect to voice communication, the extraction of information from sensor signals is a significant aspect of safety-critical applications. This requires reliable methods, while complying with applicable law. The challenge here is often the poor quality of audio signals combined with the need to process large amounts of data in real time.

In addition to the exciting scientific content shared at the conference, a rich cultural offer was also opened to the participants. The city of Charlemagne showed its best side during the various guided tours and allows many insights into its historical heritage.

The conference was organized by the Informationstechnische Gesellschaft (ITG) in VDE, in cooperation with the Technical Committee Speech Communication AT3, as well as the Technical Program Committee and the Institute for Communication Systems at RWTH Aachen University.

Extensive information can be found on the homepage of the ITG conference.

Leave a Reply

Your email address will not be published. Required fields are marked *