Media Effects Research Lab - Research Archive
Are Silent Speech Interactions the Future of Voice Activated Personal Assistants?
Student Researcher(s)
Nicholas Eng (Ph.D Candidate);
Abdelhady Abuolba (B.A. Candidate);
Faculty Supervisor
This paper was based on a project as part of the Comm 506 course.
INTRODUCTION
This study investigates differences in two different modalities of Voice Activated Personal Assistants (VAPAs). Silent Speech Interactions (SSIs) are an emerging technology that we believe could be comparable to current Voice User Interactions (VUIs) and potentially combat current consumer concerns with VUIs.
RESEARCH QUESTION / HYPOTHESES
Overarching RQ: For users of VAPAs, controlling for age and gender, what is the relationship between Silent Speech Interactions (SSI) use, and perceived ease of use, usability, usefulness, social acceptability, sense of security, privacy concern, and satisfaction?
RQ1: For VAPA users, controlling for age and gender, will SSIs evoke higher levels of sense of control, ease of use, usability, usefulness, and satisfaction, compared to VUIs?
RQ2: For VAPA users, controlling for age and gender, will SSIs evoke higher overall levels of social acceptability compared to VUIs?
RQ3: For VAPA users, controlling for age and gender, will SSIs evoke lower levels of privacy concerns and higher levels of sense of security compared to VUIs?
H1: There is a positive relationship between perceived ease of using SSIs and satisfaction.
H2: There is a positive relationship between perceived usefulness of SSI and satisfaction.
METHOD
Our research method consisted of two Adobe XD prototypes comparing the different modalities of SSIs and VUIs. Participants were recruited in a high traffic setting in a library to simulate a public place. 25 participants were randomly assigned to one of the two conditions and after completing the consent form, instructed to walk around and interact with the prototype without researcher intrusion. Participants in the silent speech condition were shown an instructional video on what the new technology is and how it works. Upon completion of the prototypes, participants were instructed by the final screen to approach the researchers to complete a questionnaire consisting of questions on their perceptions of the interaction.
RESULTS
Besides ease of use and usefulness, no statistically significant results were found for the variables of RQ1. Users in the SSI group found that interaction significantly more useful and older participants found SSIs and VUIs less easy to use. RQ2 and RQ3 were not supported with no statistically significant results being found. Ease of use was found to predict satisfaction while usability did not. Exploratory analysis was conducted finding that while controlling for gender, silent speech resulted in higher scores for usability. Along with that, a two-way interaction between gender, modality, and usability was found along with a two-way interaction between gender, modality, and satisfaction.
CONCLUSIONS/DISCUSSION
Overall, our findings suggest that if developed and released into the market, VAPAs equipped with SSI could enjoy similar levels of popularity with consumers of VUI-equipped VAPAs. Since SSIs have yet to be studied in the field of Human-Computer Interactions, our study is novel and has practical and theoretical implications for those who are in industry and academia.
For more details regarding the study contact
Dr. S. Shyam Sundar by e-mail at sss12@psu.edu or by telephone at (814) 865-2173