Virtual Rapport

Gratch, Jonathan; Okhmatovskaia, Anna; Lamothe, Francois; Marsella, Stacy; Morales, Mathieu; Werf, Rick J.; Morency, Louis–Philippe

doi:10.1007/11821830_2

Cited by 181 publications

(121 citation statements)

References 18 publications

Supporting

Mentioning

119

Contrasting

Unclassified

Order By: Relevance

“…Gratch et al (2006) had pairs of interlocutors recorded in a setting where a listener can hear the speaker's real voice while watching what they are told are graphic representations of the speaker and the speaker's gestures on monitors. In reality, the visual representation on the monitor is controlled by other parameters.…”

Section: Systemmentioning

confidence: 99%

Towards human-like spoken dialogue systems

Edlund¹,

Gustafson²,

Heldner³

et al. 2008

Speech Communication

106

View full text Add to dashboard Cite

show abstract

Section: Systemmentioning

confidence: 99%

Towards human-like spoken dialogue systems

Edlund¹,

Gustafson²,

Heldner³

et al. 2008

Speech Communication

106

View full text Add to dashboard Cite

show abstract

“…People interacting with embodied agents tend to interpret both nonverbal cues and the absence of nonverbal cues. Embodied agents can effectively communicate an intended emotion through animated facial expressions (Kätsyri and Sams, 2008 (Gratch et al, 2006). The agent used in the experiment described in this bomb experiment is shown in Figure 11 below.…”

Section: Species Design Principle 4: Encode a Response To Relay To Thmentioning

confidence: 99%

Design Principles for Special Purpose, Embodied, Conversational Intelligence with Environmental Sensors (SPECIES) Agents

Derrick¹,

Jenkins²,

Nunamaker³

2011

THCI

View full text Add to dashboard Cite

University of Arizona jnunamaker@cmi.arizona.eduAs information systems increase their ability to gather and analyze data from the natural environment and as computational power increases, the next generation of human-computer interfaces will be able to facilitate more lifelike and natural interactions with humans. This can be accomplished by using sensors to non-invasively gather information from the user, using artificial intelligence to interpret this information to perceive users' emotional and cognitive states, and using customized interfaces and responses based on embodied-conversational-agent (avatar) technology to respond to the user. We refer to this novel and unique class of intelligent agents as Special Purpose Embodied Conversational Intelligence with Environmental Sensors (SPECIES) agents. In this paper, we build on interpersonal communication theory to specify four essential design principles of all SPECIES agents. We also share findings of initial research that demonstrates how SPECIES agents can be deployed to augment human tasks. Results of this paper organize future research efforts in collectively studying and creating more robust, influential, and intelligent SPECIES agents.

show abstract

“…Gandalf analyzed in realtime an interlocutor's gaze, gesture, body stance and prosody to determine appropriate turntaking and back-channel opportunities. This has been done more recently in the Rapport Agent [13], which uses gaze, posture and prosodic perception to, among other things, detect backchannel opportunities. While performing in realtime, approaching human-level pace in some cases, none of these systems were built to adapt their behavior to their interlocutors.…”

Section: Related Workmentioning

confidence: 99%

Teaching Computers to Conduct Spoken Interviews: Breaking the Realtime Barrier with Learning

Jonsdottir

Þórisson

2009

Intelligent Virtual Agents

View full text Add to dashboard Cite

Abstract. Several challenges remain in the effort to build software capable of conducting realtime dialogue with people. Part of the problem has been a lack of realtime flexibility, especially with regards to turntaking. We have built a system that can adapt its turntaking behavior in natural dialogue, learning to minimize unwanted interruptions and "awkward silences". The system learns this dynamically during the interaction in less than 30 turns, without special training sessions. Here we describe the system and its performance when interacting with people in the role of an interviewer. A prior evaluation of the system included 10 interactions with a single artificial agent (a non-learning version of itself); the new data consists of 10 interaction sessions with 10 different humans. Results show performance to be close to a human's in natural, polite dialogue, with 20% of the turn transitions taking place in under 300 msecs and 60% under 500 msecs. The system works in real-world settings, achieving robust learning in spite of noisy data. The modularity of the architecture gives it significant potential for extensions beyond the interview scenario described here.

show abstract

Virtual Rapport

Cited by 181 publications

References 18 publications

Towards human-like spoken dialogue systems

Towards human-like spoken dialogue systems

Design Principles for Special Purpose, Embodied, Conversational Intelligence with Environmental Sensors (SPECIES) Agents

Teaching Computers to Conduct Spoken Interviews: Breaking the Realtime Barrier with Learning

Contact Info

Product

Resources

About