Change search
ReferencesLink to record
Permanent link

Direct link
Interaction of human bahaviour models (06-BRIMS-21)
KTH, School of Computer Science and Communication (CSC), Speech, Music and Hearing, TMH, Speech Communication and Technology.
2006 (English)In: Simulation Interoperability Standards Organization, 2006, 126-134 p.Conference paper (Refereed)
Abstract [en]

The field of human behaviour simulation is vast and includes intelligent behaviour of platforms as well as humanoids capable of a spoken dialogue. Implementations of human behaviour range from complex models trying to replicate perception processes, biological and cognitive functions, emotions, memory etc to simple rule based algorithms. Since the different approaches to human behaviour simulation is so different, there has been little standardization efforts in this area. A first approach to start standardization in the field of human behaviour could be to look at a human being as a black box and try to define the communication interfaces, i.e. the verbal and nonverbal communication. The interface could then be divided in three main areas: speech, visual appearance and body motion including gestures and mimic. The speech area could be compared to the simulation of radio communication in DIS and HLA RPR-FOM. In the same way as the characteristics of a radio transmitter is defined in a Transmitter PDU, we need to define the characteristics of a voice, such as whether it's a male or female voice, the pitch and speed of the voice, accents, emotional characteristics, having a cold etc. For these purposes there are a number more or less standardised protocols for voice rating, for example GRBAS (Ishiki et al, 1969) or SVEA (Hammarberg, 2000) developed for slightly different purposes by the speech therapist community. The content of the speech could be transmitted in a similar way as the Signal PDU. There are currently several standards developed for speech based man-machine communication such as VoiceXML, SALT. These standards include both speech recognition grammar and speech synthesis specifications. In computer games visual appearance is defined using Open Flight files or similar standards. However, if two human CGF's need to interact, an ontology and representation of clothing is needed in a more linguistic way. Using this representation a visualization could map this data onto an applicable Open Flight representations. Regarding body movement, there are a number of different standards. As an example, the MPEG-4 standard defines 64 points in a human face which may be altered. The representation probably needs to be at a higher level, for instance representing different emotions. Finally, the different modalities, speech, visual appearance and body motion have to be studied and described in relation to each other. Several attempts have been made for describing and standardizing human multimodal interaction. For instance, Bernsen (1997) has proposed a taxonomy for output modalities. Martin et al (2004) presented a long list of projects, initiatives and organisations that have addressed the issue of multimodal communication. This paper studies different applicable standards, which may be used to define ontology and representation for human interaction and how they could be mapped onto a Human Interaction FOM or BOMs. By doing this, interaction between different human behaviour models and humans will be simplified.

Place, publisher, year, edition, pages
2006. 126-134 p.
, Simulation Interoperability Standards Organization - 15th Conference on Behavior Representation in Modeling and Simulation 2006
Keyword [en]
Behaviour model, Communication, Interaction, Simulation, Speech, Behaviour models, Bernsen, Black boxes, Body motions, Body movements, Cognitive functions, Communication interface, Complex model, Computer game, Human behaviours, Human being, Human faces, Human interactions, MPEG-4 standard, Multi-Modal Interactions, Multimodal communications, Non-verbal communications, Output modality, Simple rules, Spoken dialogue, Visual appearance, VoiceXML, Behavioral research, Continuous speech recognition, Data visualization, Internet telephony, Motion Picture Experts Group standards, Social sciences, Speech communication, Speech synthesis, Standardization, Visualization, Computer simulation
National Category
Language Technology (Computational Linguistics)
URN: urn:nbn:se:kth:diva-155024ScopusID: 2-s2.0-84865320426ISBN: 9781604238969OAI: diva2:760282
15th Conference on Behavior Representation in Modeling and Simulation 2006, 15 May 2006 through 18 May 2006, Baltimore, MD

QC 20141103

Available from: 2014-11-03 Created: 2014-10-29 Last updated: 2014-11-03Bibliographically approved

Open Access in DiVA

No full text


Search in DiVA

By author/editor
Pakucs, Botond
By organisation
Speech Communication and Technology
Language Technology (Computational Linguistics)

Search outside of DiVA

GoogleGoogle Scholar
The number of downloads is the sum of all downloads of full texts. It may include eg previous versions that are now no longer available

Total: 6 hits
ReferencesLink to record
Permanent link

Direct link