Course Description

The central topic of this course is spoken language as it occurs in spontaneous dialogues between humans but also between humans and computers. On the one hand, we will identify the characteristics of spontaneous dialogues between humans (compared to read speech). On the other hand, we will have a look at the design of dialogue systems and discuss problems with and solutions for processing spontaneous speech.

What quantitative linguistic studies of spontaneous dialogues and investigations in the field of speech technology have in common is that they both need large corpora of spontaneous speech and appropriate methods/tools to annotate and analyze them. This course will give an introduction to tools that facilitate investigations with large speech corpora (Praat, Python), which will also already be used for student projects.

The course is suitable for both engineers and linguists. Thus, besides gaining theoretical and practical knowledge concerning dialogue systems and spontaneous speech analysis, the students will have the possibility to work in an interdisciplinary team.

Previous Knowledge
Engineers are advised to previously attend the course Linguistic Foundations of Speech and Language Technology. For linguists it might be useful to previously attend the course Natural Language Processing and/or to learn some basic programming at the beginning of the course (e.g., with tutorials provided by the lecturer).