1. How do 11 – 26 year olds currently navigate through voice assistants in the context of performing command-event-based tasks on smartphones?
2. What are user expectations and resulting design insights that can enhance user interaction with voice assistants on smartphones?
The paper discusses prompts to design voice user interfaces with better interplay between voice and touch-based commands, and auditory and visual outputs in smartphones for a seamless interaction experience.
A web-based survey designed to understand the motivation and frequency of voice assistant usage of Gen Z users on smartphones.
74 anonymised responses which were filtered to shortlist proficient Gen Z voice assistant users (in smartphones) for the next phase.
Link to survey
The goal with the interviews was to study user navigation with voice assistants on smartphones and other factors that contribute to their interaction.
Interviews were conducted over zoom in a remote setting with 8 participants.
Link to interview protocol
The first activity required the participants to pick adjectives that they associate with voice assistants.
The second activity involved participants being shown artefacts in the form of two screens with conversations between a user and a voice assistant in a given context. Participants were asked to exercise the think-aloud approach to draw insights on what the differences in the conversations indicated to them.
The prior two activities acted as the precursor to aid participants in designing their own conversation for the third and final activity.
Link to participatory design activity board
The survey was analysed on the basis of set parameters including, preference between VUIs and GUIs, frequency of engagement with voice assistants, ease of use and more.
Link to survey analysis
The interview analysis was done with parameters that built upon the ones used in the survey.
Link to interview analysis
To compliment and better understand user engagement with voice assistants, the participatory design activities helped analyse their real-time engagement within set scenarios.
Link to participatory design analysis
The literature review for the academic paper involved looking at previously conducted studies within this scope, their findings and how our study builds upon them.
Link to literature review
“......let’s say when I ask Google Assistant if my flight is on time, being presented with the flight details like the gate number on my phone screen while the assistant tells me if it’s delayed or not helps me break down the information better”
"If the voice assistant was leading me to the Amazon e-commerce app asking for smartphones of a price range, I could do that myself. So it would be better if the voice assistant was able to guide me specifically till I make my purchase based on my needs."
"Like the other day when I was traveling to "Doddaballapura," I had to stop the car and manually type the name because Siri did not understand me...it just kept repeating the same question over and over again which became frustrating."
“I would like an update from Google Assistant where it would not take up too much space as an overlay over everything else while I am doing. All I need to see is if my command is being recorded correctly. This may be a good way to get people to use it, without making it invasive.”
“The tone and personality of the voice assistant would be a huge stepping stone for me. It would be interesting if I could listen to the daily news in the voice of my favourite newsreader or listen to match highlights with the energy of a sports commentator.”
1. How might we leverage the multimodal nature of smartphone interfaces to develop better synergy between visual and sound design for voice assistant interaction?
2. How might we push the boundaries of voice assistant design beyond usability and conversation design?