Abstract

As spoken language systems become more conversational in nature, new interfaces are emerging with animated characters that are designed to elicit and manage a conversation with the user. In the present study, the spoken language of ten 6-to-10year -old children was compared while interacting with an animated character in the I SEE! interface and again while interacting with a human adult. Analyses revealed that 25% of children's utterances contained disfluencies or idiosyncratic lexical content that would be difficult for a recognizer to process. Children had significantly higher disfluency rates than adults, and a steeper slope in their rate of disfluencies as a function of utterance length. The long-term goal of this research is the development of appropriate and robust conversational interfaces for children. 1. INTRODUCTION As spoken language systems gradually have evolved toward more conversational exchanges, parallel interest has grown in developing interfaces with animated char...

Citations

84	Multimodal Interactive Maps: Designing for Human Performance - Oviatt - 1997
63	Modeling hyperarticulate speech during human-computer error resolution - Oviatt, Levow, et al. - 1996
47	Predicting Spoken Disfluencies during HumanComputer Interaction,” Computer Speech and Language - Oviatt - 1995
20	Designing and evaluating conversational interfaces with animated characters - Oviatt, Adams
19	A study of speech recognition for children and the elderly - Wilpon, Jacobsen - 1996
15	The development of communication skills: Modifications in the speech of young children as a function of listener - Shatz, Gelman - 1973
9	Improvements in children’s speech recognition performance - Das, Nix, et al. - 1998
4	How Effective is Unsupervised Data Collection for Children's Speech Recognition - Aist - 1998