NLP / Dialogue Classes at Bielefeld University, Summer 2013

The classes I will be teaching this summer at Bielefeld University:

  • advanced methods in applied computational linguistics 2; MA; basically covers machine learning methods for classification (the prequel covers sequence classification and graphical models). Will feature heavy use of IPython notebooks, scikit-learn, and nltk.
  • natural language processing; BA; will mostly be about parsing and grammar, & will use nltk.
  • practical dialogue modelling; MA; small graduate class, where we will look at existing dialogue managers and implement a small toy system.

Interspeech 2013 Tutorial on Incremental Processing

Timo Baumann and I will be giving a tutorial on “recent advances in incremental spoken language processing” at this year’s interspeech in Lyon. We’ll create a separate page for that later, but for now, here’s the short description:

Incremental processing — that is, the processing of user input while it is still ongoing, and the preparation of possibly concurrent system reactions — is about to make the move from a research challenge to being deployed and beneficial for users (e.g. in Google Voice Search). Beyond speeding up the presentation of results of voice search, it offers the potential for creating spoken dialogue systems with a much more natural behaviour, with respect to turn-taking or production and understanding of feedback utterances.

In this tutorial, we will discuss the challenges posed by incremental spoken language processing, present the state of the art in incremental processing, and in particular will describe and demonstrate a framework for incremental processing that we have implemented over the last years which offers an architecture for creating systems by connecting modules, and contains reference implementations of the full chain of modules: ASR, NLU, Dialogue Management, Action Selection, Natural Language Generation, Speech Synthesis.

The tutorial is targeted at researchers interested in incremental processing in general, and in particular at researchers who are interested in “incrementalizing” individual processing modules (such as natural language understanding, dialogue managements, natural language generation, etc.) and want to be able to quickly realise end-to-end systems within which to test their modules.

Recent Talks: Honda Research Institute, Tokyo; ESSV 2013

Earlier this month, I visited my good colleagues Mikio Nakano and Kotaro Funakoshi at Honda Research Institute Tokyo. I got some very interesting demos of the stuff they are working on at the moment, and I also gave a talk to the group there about our recent work (basically, the 2013 version of my “incremental dialogue processing” talk). Earlier this week now, I gave a shorter version here in Bielefeld at the “Elektronische Sprachsignalverarbeitung” conference (German signal processing conference). Without the animations and videos the slides look a lot less interesting (really, believe me, they are great when things are flying around!), but I’ll still put them here as they give a nice overview of the current state of our work.

The slides briefly mention at the end our (multimodal interaction analysis tools), our toolset for running and analysing multimodal experiments (and integrating multimodal data streams into interactive systems). I will write about this in more detail soon; we’re also preparing a static page on that.