Andreas Liesenfeld


interactional linguistics & speech technology

Modeling action formation

Understanding action formation

Framenets and Constructicons

Machine learning for pragmatics

Video corpus: MYCanCor

Tools, tutorials and links

Find me on Twitter

Find me on Google scholar

★★★ Upcoming: Looking forward to presenting at iPra 2019 in Hong Kong, 9 June 2019. ★★★

Picture I am currently a PhD candidate in linguistics at Nanyang Technological University, Singapore. A linguist with a professional background in data science, I study talk with a focus on understanding the interplay between social action and linguistic structure. I am also developing new tools for conversational (nontask-oriented) dialog systems. I have worked on annotations of non-lexical utterances, disfluency and laughter for the application in spoken language understanding (SLU). I have also worked on the development of cognitive grammars for conversational systems, i.e. framenets and constructicons.


My main research interest is the study of talk, combining qualitative studies of talk-in-interaction (interactional linguistics, conversation analysis) with the quantitative study of corpora of spoken discourse (grammar induction, distributional modeling). In my early-career research I have conducted linguistic fieldwork in different parts of Asia and built a video corpus of Malaysian Cantonese. Currently, I work on developing corpus-based representations of building blocks of conversation as tensors. Specifically, I explore how speech corpus data can be integrated in natural language understanding modules.

Most of my work has been on Chinese (Mandarin and Cantonese) and English (British Englishes and World Englishes).

PhD student in linguistics, Nanyang Technological University, Singapore
Visiting fellow, CRC991 frame semantics, Düsseldorf University, Germany
Data scientist, Industry, Berlin, Germany
Master, School of Social Sciences, University of Hong Kong, SAR of China
Bachelor, Fakultät für Geisteswissenschaften, University of Bonn, Germany
Get full CV

Visit Google scholar
For the video corpus of spoken Chinese see also: MYCanCor

Cause Chi (Tagger for causal discourse markers in spoken Chinese)
MV parser (Augument structure tagger for Chinese)

Conversation analytic transcription in LaTeX
Multilayed phonetic annotation with Praat

♦ ♦ ♦ I recently moved this blog from to github and it'll take some time to copy all the content over from wordpress to jekyll. ♦ ♦ ♦