Abstract

Due to the increase in virtual humans as a pedagogical agent, this study investigates how perceptions of virtual humans are affected by voice quality in understanding a budgeting scenario where trust is essential to learning and application. Eighty-six participants were randomly assigned to three conditions where voice quality for a virtual human varied (human voice, low quality text-to-speech, high quality text-to-speech) when narrating a financial literacy course. Measures of trust and course comprehension were collected. Results of the learning assessment suggest no difference in comprehension based on voice quality of a virtual human. No differences were observed between the voice quality groups in participants' perception of trust, the abilities to facilitate learning, credibility of the agent, human-likeness of the agent, or how engaging the agent was. This is possibly due to the lower age demographic who have become increasingly exposed to virtual human voices through popular platforms such as Tik-Tok.).

Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call