How we trust, perceive, and learn from virtual humans: The influence of voice quality

Chiou, Erin K.; Schroeder, Noah L.; Craig, Scotty D.

doi:10.1016/j.compedu.2019.103756

Cited by 43 publications

(32 citation statements)

References 47 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…According to the voice principle, spoken sentences should be pronounced with normal intonation and not with a mechanically distorted voice or a foreign language accent. It is explained analogously to the personalization principle (Mayer 2014b ) and supported by several experiments (e.g., Liew et al 2020 ; Chiou et al 2020 ). Mayer et al ( 2003 ), for example, conducted two experiments on lightning showing that an audio commentary presented in a conventional accent instead of a foreign language accent (experiment 1) or a machine voice (experiment 2) improved the transfer learning performance and led to more positive evaluations of the speaker.…”

Section: Social Cues In Dynamic Learning Materialsmentioning

confidence: 54%

“…Besides cultural differences, previous studies showed a speakers’ enthusiastic voice leads to higher social and affective ratings and an increased transfer performance compared to a calm voice (Liew et al 2020 ). Also, a high-quality voice (made with a highly elaborated speech engine) leads to higher ratings of credibility and engagement, compared to a low-quality voice (Chiou et al 2020 ). Recent research further outlines the importance of the quality of the voice on perceived trust in the online learning environment.…”

Section: Social Cues In Dynamic Learning Materialsmentioning

confidence: 99%

See 1 more Smart Citation

The Cognitive-Affective-Social Theory of Learning in digital Environments (CASTLE)

et al. 2021

View full text Add to dashboard Cite

For a long time, research on individuals learning in digital environments was primarily based on cognitive-oriented theories. This paper aims at providing evidence that social processes affect individual learning with digital materials. Based on these theories and empirical results, a social-processes-augmented theory is suggested: the Cognitive-Affective-Social Theory of Learning in digital Environments (CASTLE). This CASTLE postulates that social cues in digital materials activate social schemata in learners leading to enhanced (para-)social, motivational, emotional, and metacognitive processes. To substantiate this theory, socio-cognitive theories are used, which predict social influences on learning with digital materials. Besides, previous empirical findings are presented assuming that with a rising number of social cues in digital materials, the influence of social processes increases. Finally, consequences regarding the design of digital learning media are discussed.

show abstract

Section: Social Cues In Dynamic Learning Materialsmentioning

confidence: 54%

Section: Social Cues In Dynamic Learning Materialsmentioning

confidence: 99%

The Cognitive-Affective-Social Theory of Learning in digital Environments (CASTLE)

et al. 2021

View full text Add to dashboard Cite

show abstract

“…Similarly, Abdulrahman et al [2] found that human and synthetic voices were equally good at reducing feelings of stress. In their respective studies, Davis, Vincent, and Park [42] and Chiou et al [27] found that there was no difference for learning measures. Sims et al [166] found that when viewing a robot needing assistance, people were more likely to give commands when it had a synthetic voice rather than a human voice, suggesting that the voice could mediate whether people treat a robot as a capable human or a machine that needs direction.…”

Section: Anthropomorphism Humanlikeness and Natural Vs Synthetic Voicesmentioning

confidence: 95%

“…Chérif and Lemoine [25] found that human voices elicited greater social presence compared to synthetic voices in a website, but there was no effect on perceived trustworthiness. Chiou et al [27] found that the human voice was the most trusted and engaging, and human as well as high-quality synthetic voices were most credible. Craig and Schroeder [32] found that human and modern TTSs were superior to older TTSs in terms of perceived credibility and engagement.…”

Section: Anthropomorphism Humanlikeness and Natural Vs Synthetic Voicesmentioning

confidence: 99%

Voice in Human–Agent Interaction

Seaborn

Miyake

Pennefather³

et al. 2021

ACM Comput. Surv.

105

View full text Add to dashboard Cite

Social robots, conversational agents, voice assistants, and other embodied AI are increasingly a feature of everyday life. What connects these various types of intelligent agents is their ability to interact with people through voice. Voice is becoming an essential modality of embodiment, communication, and interaction between computer-based agents and end-users. This survey presents a meta-synthesis on agent voice in the design and experience of agents from a human-centered perspective: voice-based human--agent interaction (vHAI). Findings emphasize the social role of voice in HAI as well as circumscribe a relationship between agent voice and body, corresponding to human models of social psychology and cognition. Additionally, changes in perceptions of and reactions to agent voice over time reveals a generational shift coinciding with the commercial proliferation of mobile voice assistants. The main contributions of this work are a vHAI classification framework for voice across various agent forms, contexts, and user groups, a critical analysis grounded in key theories, and an identification of future directions for the oncoming wave of vocal machines.

show abstract

“…According to the voice effect theory (Craig and Schroeder, 2017 ), compared with the voice synthesized by computers, the way of using instructor's voice recording in the video course was more able to promote learner's deep learning and enhance learning performance. Although instructor voice recording is more in line with the learner's preferences, with the development of voice synthesis technology, the difference between computer-synthesized voice and the human voice is rapidly shrinking, and it has broad future development prospects (Chiou et al, 2020 ).…”

Section: Introductionmentioning

confidence: 99%

Would It Be Better if Instructors Technically Adjust Their Image or Voice in Online Courses? Impact of the Way of Instructor Presence on Online Learning

et al. 2021

View full text Add to dashboard Cite

This study investigated the effects of the changes in the image and voice of instructors in online video courses on online learner's learning achievement, social presence, learning satisfaction, and academic emotion. Two simultaneous online experiments were conducted with 122 college students in the image experiment, where the course videos varied in terms of the instructor's image (original image, face-beautified image, virtual image, and no image), and 93 college students in the voice experiment, where the course videos varied in terms of the instructor's voice (original voice, mutated voice, computer-synthesized voice). The results showed that learners viewing videos without instructor images had better learning achievements and less academic boredom relative to those who viewed videos with instructor images. However, the real instructor images were able to promote learners' learning satisfaction of instructor-student interaction more than no image and virtual image and promote satisfaction of instructor teaching more than virtual image. Meanwhile, learners' evaluation of the real instructor images was better than that of the virtual instructor image, and their evaluation of the face-beautified instructor image was better than the original image. Moreover, learners evaluated real instructor voices better than the computer-synthesized voice. In addition, the linear regression analysis revealed that the evaluations of both instructor's image and voice had a positive relationship with learners' social presence, learning satisfaction, and enjoyment, whereas they had a negative relationship with learner's boredom. And the evaluation of the instructor's image positively predicted student's transfer learning achievement. Thus, we suggested that the way of instructor presence should be well-designed and integrated with the course's instructional design and image and voice processing technology can be applied to assist online video course development.

show abstract

How we trust, perceive, and learn from virtual humans: The influence of voice quality

Cited by 43 publications

References 47 publications

The Cognitive-Affective-Social Theory of Learning in digital Environments (CASTLE)

The Cognitive-Affective-Social Theory of Learning in digital Environments (CASTLE)

Voice in Human–Agent Interaction

Would It Be Better if Instructors Technically Adjust Their Image or Voice in Online Courses? Impact of the Way of Instructor Presence on Online Learning

Contact Info

Product

Resources

About