InBetween

CONTEXT

InBetween explores how video calls can show that someone may be preparing to speak. Participant tiles reveal more of the video as cues appear, then open fully when speech is detected. Interviews and two prototype rounds shaped this distinction between possible readiness and actual speech.

ROLE

Research, UX/UI and interaction prototyping

I led research framing and 12 interviews, designed the interface and cue logic, built live prototypes, and evaluated the design with participants.

PROBLEM DISCOVERY

Finding a moment to speak was difficult to judge.

I interviewed 12 people about taking turns in video calls. Participants described waiting because someone else might speak at the same time. Active-speaker indicators helped identify the current speaker, but offered little information before speech began.

PRECEDENTS

Existing tools leave a gap before someone speaks.

I compared four ways meeting platforms support participation. They identify the current speaker, register a request, or offer another channel for responding. This led me to explore a tentative cue within the participant tile, before a formal hand raise or detected speech.

PROTOTYPE ITERATION

From abstract shapes to participant video.

Two prototype rounds, each with 13 participants, changed the design in different ways. The first informed the visual representation; the second exposed false activations in the input logic.

INTERACTION SYSTEM

Partial reveal for readiness. Full reveal for speech.

Listening keeps the video in a narrow strip. Unmuting creates a small reveal; mouth movement combined with gaze or leaning creates a larger one. Detected speech opens the full view.

ADDITIONAL EXPLORATION

Exploring a response to shared silence

A separate prototype explored comments about awkward pauses. After five seconds without speech, Circle Mode slowly rotates the participant tiles. This exploration was not part of the results below; whether the movement supports conversation or distracts from it remains to be tested.

EVALUATION

Evaluating clarity and timing in conversation

In a short group conversation task, 13 participants compared a baseline video grid with the live reveal prototype. They rated how clearly they could recognize an upcoming speaker and perceive others’ speaking intention on a five-point scale.

Experience film

TAKEAWAYS

What I learned from designing Inbetween

Testing changed how I represented readiness and how I decided when the cue should appear.