CASE STUDY

VR SPEAK

PUBLIC SPEAKING TRAINING

Building Confidence in Early Career Professionals, with VR Public Speaking Training Platform

Role : VR UX Designer & Developer

INTRODUCTION

About the Project

VR Speak is a level-based Virtual Reality platform designed to help users build communication confidence through immersive, progressive speaking scenarios. Developed as part of a graduate HCI research lab, the project explores how VR can reduce speech anxiety using UX research and gamified design.

GET STARTED

Overview

Challenge:
How might we use immersive technology to help people overcome their fear of public speaking in a safe, structured, and engaging way?

Approach:
We created a VR training tool with three levels of difficulty (Café → Classroom → Auditorium), using user research to inform interaction design, pacing, and feedback mechanics.

Outcome:
A working VR prototype tested with users, with early feedback showing increased self-reported confidence and strong usability ratings.

The Story

The project started from a personal insight—many peers feared speaking in public, even in low-stakes settings. Existing training solutions felt too intimidating, rigid, or inaccessible. We hypothesized that a gamified VR experience could offer a safer space for users to practice speaking, fail privately, and build confidence gradually.

I initiated user interviews, mapped emotional pain points, and defined experience principles like "Safe Exposure" and "Progressive Challenge" to guide our design.

| Problem Space

77% of adults report anxiety around public speaking.

  • Existing solutions (like Toastmasters) lack low-pressure environments for true beginners.

  • Most VR tools drop users into high-stress simulations without onboarding or progression.

  • There’s an unmet need for confidence-first communication training.

This gap became our focus: design a VR tool that builds confidence, not just performance.

Target User

Early career professionals

People with speech anxiety

Non-native English speakers

Professionals in client-facing roles

Design Citeria

Structured learning flow

Emotional safety

Minimal cognitive load

Intuitive navigation

| Glimpse of Solution

MODES

User can select the type of mode they want to play in - Practice ( relaxed )and Game ( challenging ).

AUDIENCE

Users face a realistic audience ( 360 video) , with varied reactions (engaged, unintrested).

CONFIDENCE RATING

Users rate their performance after delivering the speech, which helps them see visible self growth.

RESEARCH

| Interview Insights

We conducted 8 semi-structured interviews with students and early professionals who self-identified as anxious public speakers. Our goal was to unpack their emotional and behavioral patterns around communication anxiety.


These insights grounded our hypothesis: VR could provide a psychologically safe practice zone with enough realism to trigger growth, but without the social pressure of real audiences.

“I worry more about being judged than forgetting my words.”

Many avoided speaking tasks altogether, leading to confidence erosion over time.

Traditional public speaking training felt too performative and not personal enough.

Users craved private practice spaces where failure felt safe and invisible.

UNDERSTANDING USERS

| Empathy Mapping

We synthesized interview data into an empathy map to humanize the problem space and uncover emotional truths beyond behaviors.


This exercise helped us shift focus from just training performance to designing for emotional resilience.

DECISION MAKING

| Game Design Strategy

We leaned on gamification and level design principles to shape engagement:

  • Progressive difficulty: Users start in a calm café setting, then move to a classroom, and finally a high-pressure auditorium.

  • Feedback mechanics: Simulated audience reactions (distraction, interest, applause) gave users real-time cues without verbal judgment.

  • Safe failure loops: Users could retry scenarios, re-record their speeches, and view progress indicators.

DESIGN AND ITERATION

| Prototyping

We followed a lean, iterative prototyping approach, starting with low—fidelity 3D prototyping, evolving to mid fidelity working Unity VR prototype and finally building a high-fidelity Unity VR prototype.

LOW FIDELITY : 3D PROTOTYPE

We created a low-fidelity 3D prototype to visualize the gameplay, user flows, and in-game interactions. The scenarios were created using clay, paper pieces, and background images.

MID-FIDELITY : VR UNITY PROTOTYPE

After finalizing the user flow, we began developing the VR prototype in Unity. We incorporated three different 360-degree video scenarios to simulate audience reactions: a casual conversation with two friends, a small group presentation in front of 8–10 people, and a formal speech delivered to a full room. Each scenario concludes with users submitting a self-assessment of their performance to help track and reflect on their confidence levels.

HIGH-FIDELITY : VR UNITY PROTOTYPE

After testing the mid-fidelity prototype with 10 users, we developed the high-fidelity version by incorporating gameplay adjustments based on user feedback to enhance usability and overall user experience.

TESTING

| User Testing

After building the Unity prototype of VR Speak, we conducted a round of formative usability testing with 10 participants — all of whom had previously identified themselves as experiencing some level of public speaking anxiety. Our goal was to evaluate:

  • Usability of the onboarding and interface

  • Emotional response across levels

  • Perceived realism and effectiveness of VR settings

  • Impact on self-reported communication confidence


Testing Setup

  • Participants: 10 individuals with self-reported public speaking anxiety

  • Prototype: Mid-fidelity Unity build with 3 levels: Café, Classroom, Auditorium

  • Tasks: Navigate through selected VR scenarios, engage in speech tasks, then self-rate and reflect

  • Methods: Observational walkthroughs + post-session interviews

Key Insights

  • 1. Clarity Drives Confidence

    Users struggled when prompts or objectives weren’t explicit—especially in the Café level. Lack of clear direction led to hesitation or silence.


    “I felt most unsure in the first level since there were no prompts on when to speak.”


    Takeaway: Even in low-pressure environments, users need scaffolded guidance (e.g., audio cues, character prompts) to feel empowered to begin.

  • 2. Scenario Tone > Audience Size

    Contrary to our hypothesis, user comfort wasn't directly tied to audience size. Some felt more confident in structured classroom scenarios than in the casual café scene.


    “The third scenario, that was a classroom, was helpful because the people seemed helpful... So that was comfortable.”


    Takeaway: Comfort is shaped more by emotional tone and perceived expectations than crowd size. Realistic but supportive environments build more trust.

  • 3. Presence Heightens Realism—and Anxiety

    The more immersive the environment, the more users responded with authentic emotional and physical behaviors (e.g., avoiding eye contact, adjusting posture).


    “I was definitely present... I can just look away from the people.”


    Takeaway: Realism in VR evokes genuine reactions, which makes it a powerful tool—but it requires emotional pacing and safety mechanisms to avoid overwhelming users.

  • 4. Users Crave Real-World Complexity

    Participants wanted scenarios that mimicked real social nuances: interrupting, joining mid-conversation, or answering spontaneous questions.


    “A scenario that can help would be there’s an ongoing conversation that you have to join…”


    Takeaway: Beyond speech delivery, users are eager to practice unscripted, spontaneous social interactions—key for real-world confidence.

What We Learned

User testing revealed that building confidence is not just about exposure—it’s about creating emotionally intelligent environments. Users respond best when they know what’s expected, feel supported, and are given room to grow without pressure.

Confidence is built not by asking users to “perform,” but by letting them safely try, reflect, and improve.

CHALLENGES FACED

| Challenges

  • Recruitment barriers: People are reluctant to talk about social fears. We adapted by offering anonymous prompts and empathy-based framing.

  • Realism vs safety: Striking a balance between immersive pressure and emotional safety required careful environment design and pacing.

  • Prototyping constraints: Translating design intent into Unity with limited VR development experience pushed us to prioritize core experience over fidelity.

  • Measuring success: We chose qualitative methods (user journaling, reflection prompts) to evaluate confidence changes ethically.

KEY TAKEAWAYS

| Takeaways

Progressive exposure matters. Confidence is best built in steps, not leaps.

  • Designing for emotion requires research-backed intuition. Empathy maps, not just personas, helped us capture the real anxieties behind behavior.

  • Cross-functional prototyping sharpens decisions. Working across design, research, and dev let us evaluate tradeoffs quickly.

  • Impact isn't always visible—until it is. A single user saying, “I could finally speak without freezing up,” validated the months of effort.


KEY TAKEAWAYS

| Reflection

This project pushed me to design for emotional outcomes, not just functional success. Here’s what I took away:

What I Learned:

  • Confidence is experiential. You can’t design it directly—you have to create the conditions for it to grow.

  • VR demands nuance. Designing for immersion means balancing realism with safety, and always watching for cognitive overload.

  • User feedback is not always verbal. In VR, posture, hesitation, and repeat attempts often told us more than words.


What I’m proud of:

  • Staying anchored in research while navigating new tech

  • Designing a product that users described as “comforting” and “empowering”

  • Translating a human problem into a tangible, interactive solution