Evaluation of avatar and voice transform in programming e-learning lectures
- PDF / 1,819,522 Bytes
- 9 Pages / 595.276 x 790.866 pts Page_size
- 115 Downloads / 177 Views
ORIGINAL PAPER
Evaluation of avatar and voice transform in programming e-learning lectures Rex Hsieh1
· Hisashi Sato1
Received: 9 September 2019 / Accepted: 13 October 2020 © Springer Nature Switzerland AG 2020
Abstract This article reports the effectiveness of high frame rate facial animated avatar and voice transformer in eLearning. Three avatars: (real male professor, male avatar, female avatar) were combined with male professor’s voice or VT-4 vocoder transformed voice to create six distinguished videos which were then viewed by university freshmen students. A total of 186 students divided into 15 groups participated in this experiment. Female avatar was the most appealing avatar visually, but its combination with voice transform severely hinders its overall score. This research can be extended to real time live evaluation measuring preferences of students and draw more connections between student perception of avatar and actual lecturers. Keywords eLearning · Avatar · User experience · Voice transform
1 Motivation The advancement in technology brought about the introduction of eLearning to educational institutes. By supplementing traditional courses with eLearning materials, instructors are able to introduce new learning methods without completely deviating from standard education programs [3,5]. Some of the most popular forms of eLearning include online courses, video clips of lectures, and gamification of courses and materials [2]. The motivation behind developing eLearning platforms includes enabling students to learn anywhere anytime [1]. The increase in eLearning system has also led to an increase in literature evaluating the effectiveness of various eLearning components. One such study was by Craig and Schroeder [6] which utilized three different audios: classic and modern voice engine produced voice and human voice, to measure the effectiveness of each audio have on learning. The study does not measure the effectiveness of altered human voice on learning and how audio when paired with avatars
B
Rex Hsieh [email protected] Hisashi Sato [email protected]
1
Department of Information Media, Kanagawa Institute of Technology, 1030 Shimoogino, Atsugi, Kanagawa 243-0292, Japan
of different genders can have different effect on learning; however, which is what this research seeks to answer. Other research such as one by Walsh [7] tries to teach health care using eLearning contents including video but did not include a variety of avatars and audios. While some past research have argued that avatar styles and whether naturalism or stylization is better as a pedagogical avatar [8], thankfully this really is not an issue in Japan as the anime style is a widely recognizable style that is immediately accessible to the Japanese audience therefore bypasses the question of naturalism vs stylization in favor of a widely accepted style. Furthermore, while some previous research such as one by Baylor et al. [12] found that avatars sharing the same gender with the viewers were able to yield more influence, based
Data Loading...
