The 1st Workshop on Child, Computer and Interaction (WOCCI2008)

Chania, Crete, Greece
October 23, 2008

Detecting Problems in Spoken Child-Computer Interaction

Dino Seppi (1), Matteo Gerosa (1), Björn Schuller (2), Anton Batliner (3), Stefan Steidl (3)

(1) Fondazione Bruno Kessler, irst, Trento, Italy
(2) Institute for Human-Machine Communication, Technische Universität München, Munich, Germany
(3) Lehrstuhl für Mustererkennung, Friedrich-Alexander-Universität Erlangen, Germany

In this paper we describe the effectiveness of some linguistic features for detecting problems in spoken child-computer interactions. To this aim, we use an Automatic Speech Recognizer for generating the spoken word chain, and a word tokenizer for obtaining the lexical and stemming information. Automatic classification of each turn is eventually achieved by exploiting the frequencies of tokens’ classes. The impact of ASR and tagger accuracy on automatic detection are discussed by comparing fully automatic with manually corrected approaches.

Full Paper

Bibliographic reference.  Seppi, Dino / Gerosa, Matteo / Schuller, Björn / Batliner, Anton / Steidl, Stefan (2008): "Detecting problems in spoken child-computer interaction", In WOCCI-2008, paper 15.