User representations of computer systems in human-computer speech interaction

Amalberti, René; Carbonell, Noëlle; Falzon, Pierre

doi:10.1006/imms.1993.1026

Cited by 95 publications

(60 citation statements)

References 0 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…2.1 of [1]. The findings stemming from the analysis of the verbal protocols collected during E1 contradict these results: we observed no significant statistical difference between the linguistic behaviors and expression styles of the Gexp and Gref groups [1].…”

Section: Other Drawbackscontrasting

confidence: 83%

“…The main objective of the first study [1] was to determine the possible influence of the dialogue context -human communication versus human-computer interaction -on the verbal expression, dialogue behavior, and problem solving strategies of potential users. This study is based on a complex hybrid approach involving empirical techniques within an experimental overall framework.…”

Section: Objective and Overall Setupmentioning

confidence: 99%

See 1 more Smart Citation

Towards the design of usable multimodal interaction languages

Carbonell

2003

Universal Access in the Information Society

View full text Add to dashboard Cite

UAIS. Article dans revue scientifique avec comité de lecture. The original publication is available at www.springerlink.comInternational audienceThis paper presents novel recommendations for the design of usable multimodal command or query languages. These recommendations have been inferred from the results of three empirical studies focused on the use of spontaneous speech (first study) and the synergic use of spontaneous versus controlled speech and gestures (second and third studies) for interacting with current application software. In particular, we propose a method for designing multimodal languages which can be considered as appropriate substitutes for direct manipulation, in all contexts precluding the use of mouse and keyboard, and for all standard categories of users, especially the general public. || Nous présentons de nouvelles recommandations pour la conception de langages d'interaction multimodaux robustes, faciles à apprendre et à utiliser. Ces recommandations ont été établies à partir des résultats de trois études empiriques sur l'usage spontan

show abstract

Section: Other Drawbackscontrasting

confidence: 83%

Section: Objective and Overall Setupmentioning

confidence: 99%

Towards the design of usable multimodal interaction languages

Carbonell

2003

Universal Access in the Information Society

View full text Add to dashboard Cite

show abstract

“…Evidence suggests that users' behaviour is influenced by their experience with a system as well as by their initial expectations of it. Amalberti et al (1993) found differences between a group that thought they were talking to a computer and one that thought they were talking to a human over a noisy channel (which was in reality the case for both groups). However, the differences they were mostly present at the beginning of the interactions and dissipated with time.…”

Section: Design Implicationsmentioning

confidence: 96%

Towards human-like spoken dialogue systems

Edlund¹,

Gustafson²,

Heldner³

et al. 2008

Speech Communication

106

View full text Add to dashboard Cite

show abstract

“…The influence of the interlocutor model, which can be defined as "…the representation of the interlocutor's technical and linguistic skills, which is partly constructed during the dialogue process" (Amalberti et al, 1993, p.551), extends beyond human-human dialogue, as individuals may profoundly modify the way they formulate their message according to the type of interlocutor that is being modelled (system or human operator). Amalberti et al (1993) found that the number of relevant items of information contained in an initial speaking turn was lower in a group interacting with an (simulated) artificial system than in a group interacting with a human operator. Users' knowledge (erroneous, in this case) of the system's understanding and problem-solving capabilities led them to provide fewer items of useful information in their initial request when they interacted with the system rather than with a human operator.…”

Section: Impact Of Users' Representations Of a System On Their Utteramentioning

confidence: 87%