An interactive memory log for individuals learning their partner's first language.


Advised by Deloitte Digital.

Timeline
Mar - Jul 2024

Team
2 Designers
2 Researchers

What I delivered

User Research & Synthesis

Conducted dyad user interviews, synthesized findings and produced research report.

-->

UX Design Iterations

Oversaw ideation and storyboarding workshops, wireframed user flows and prototyped final mockups.

-->

Design Systems

Developed style guide and component library to ensure consistency across the design phase.

-->
●  To Final Design  ->

01. Final Design

Introduce Crumbz.

Capturing Moments

Crumbz automatically detects when the target language is in used. With a simple tap of the fingers, users can start recording or make a timestamp.

Accessing Stamped Moments On-the-Go

Navigate to directly any stamped moment when the memory is ready for review.

Chatlog, Stamped Moments, and Memory Cards

Navigate from homepage to the captured memory to see full transcript, time stamps, or bite-sized cards.

Memory cards

Learn language concepts through curated learning cards and explore relevant cultural facts and contextual use casesdy for review.

Desktop: Structured Learning

Desktop: Explore Memories

02. Research & Scoping

Problem: Language Learning for Multiligual Couples.

Our literature review revealed a gap in common language learning tools, which often fall short in addressing the needs of individuals learning their partner's first language, particularly when it comes to navigating cultural nuances and maintaining authentic communication in a multilingual setting.

Semi-structured Dyad Interviews

Talking to Multilingual Couples Engaged in Language Learning

We conducted 90-minute interviews with 8 couples where one partner is learning the other's native language. Interviewing both parties together allowed us to observe their interactions and gather behavioral data in order to gain insights into our research questions:

1.  How do multilingual romantic relationships affect the experience of language learning?

2.  How does the one learn their partner’s language in a multilingual romantic relationship?

3.  What role does the native speaker in the relationship play in LLE of the learner?

●  Data Affinity Map  ->

Key Findings

Learners value the effectiveness of structured learning but often resort to more flexible learning methods due to limited time and energy.

"I’m thinking very heavily about going to actual classes, I feel like that was the best way for me to learn. But the problem is, I don’t have that much time."—  Learner 3

The partners complement the learners’ primary learning material by providing cultural and situational contexts.

"Sometimes he's like, okay, this is not what our teachers told us in class. I'm like, well, that's how we say it in daily life."—  Partner 5

Couples face challenges in balancing communication and learning, often resorting to translation and disrupting the natural flow of conversations.

"Because they (L3 and P3’s parents) don't understand each other. They're literally talking to me at the same time. So I get very frustrated."—  Partner 3

Language learning deepens couple's connection by creating delightful moments and memorable interactions.

"When L7 gave a speech at our second wedding reception, all of my family understood it. It also speaks to a deeper part of me."—  Partner 7

Design Principles

Scoping the Problem

Based on the key findings, we developed three design principles to guide the design phase and help evaluate design concepts.

Build a safe space where learners feel comfortable making mistakes.

Celebrate delightful and memorable moments of everyday life.

Deepen emotional and cultural connections between the couple and the families

●  Full Research Report  ->

03. Ideation, Design, Evaluation

Crafting a Holistic Learning Experience.

Ideation

Exploring Concepts

After generating 40 concepts exploring different user flows and form factors, we affinitized them around themes, allowing the team to evaluate groups of ideas against our design principles.

Standout ideas: an interactive textbook, a shared memory journal, and a smart conversation log leveraging everyday conversations for learning.

Synthesizing Ideas

I facilitated a downselection workshop, leading the team to align on components from different concepts tha best address our design principles. These elements became the building blocks of our final design.

Final Concept

Our final design is an interactive memory log that enables learners to capture conversations, add real-time timestamps, and later review and study the language within the context of those authentic interactions.

I created these concept sketches for internal crit sessions.

User Flow

Sequencing the Interactions

I translated the chosen concept into a detailed user flow diagram. The primary challenge was that our idea spans multiple touchpoints—from capturing conversations to enabling both short, flexible study sessions and longer, more structured ones. Each touchpoint required a distinct interface, making it crucial to ensure seamless navigation across various devices.

● mobile interface

● Smartwatch & desktop interface

Wireframing

Prototyping the Interactions

Given the multiple touchpoints across different devices, I collaborated with another designer to wireframe the interactions. I designed the prioritized mobile interactions to support short, flexible study sessions while allowing users to transition back to more structured study formats. This prototype would be used for concept testing with a group of recruited users.

Storyboarding

Contextualizing the Experience

Following the low-fidelity prototypes, we created a storyboard to bring the interfaces into real-life settings. The storyboard served two purposes:
1. Communicate the envisioned experience with the product to users during concept testing, and second
2. Align our team internally, ensuring our decisions remained centered around the users.

Concept Testing

Getting Feedback From Users

After developing the storyboard, we conducted five 1-hour concept test sessions to assess the overall desirability and effectiveness, identify barriers in user flows, and uncover areas for improvement. We sought feedback on how the prototype made participants feel and whether they believed it could support language learning and deepen their connection with their partner.

Feedback A: Disconnection Between Main Features

Participants appreciated the short, flexible learning experience supported by learning cards but found the user flows to be somewhat disconnected. They questioned how users would navigate between timestamped moments, full chat logs, and learning cards. Additionally, they wanted the ability to highlight and convert selected content into learning cards.

Update A.1: Segmented Control
I reorganized the core learning features into a segmented control within the main conversation view. This allows for simple navigation between different study modes.

Update A.2: Memory Card Backtracking
In the memory card interface, I added options for users to navigate back to the full list of cards or directly to the chatlog moment where the word was used.

Update A.3: Custom Memory Cards
In the full chat log or at any timestamped moment, I added a highlight feature for users to add specific vocabulary or cultural concepts to their collection of memory cards.

Feedback B: Consent and Recording Concerns

There was noticeable discomfort regarding the recording feature. Participants desired a more articulate process for managing consent, including clearer options for opting out and understanding how opting out would impact the recording and overall functionality.

Update B.1: Clarified Consent Process
We adjusted the consent prompt for clearer communication with conversation participants. Individuals who have previously granted consent will not need to do so again but can still opt out of the recording, even after the conversation has ended.

Update B.2: Future-Facing Infrastructure
We clarified the consent process such that the memory capturing feature now leverages Bluetooth Low Energy to detect nearby individuals and Voice Separation technology to isolate specific voices. We envision a protocol built into all personal devices, allowing people to manage their consent status for all digital applications.

Design Systems

I built the design system to streamline the team's workflow and allow the team to focus on the user flows during design iterations. The system also ensured cohesiveness across different touchpoints, each tailored to a different device interface.

Reflection

This product is designed to complement traditional learning resources like textbooks, language classes, and apps. While those tools offer a systematic approach to grammar and vocabulary, our product provides the real-world application and cultural context often missing from conventional resources. It's important to recognize that this product is not a standalone solution for language learning. I hope to continue to engage with user community and conduct further tests and workshops to understand how it fits into the broader language-learning journey.

The broad focus if Crumbz—covering features like recording conversations, teaching language, exploring culture, and providing context—poses a risk of diluting its core purpose. Moving forward, I plan to establish priority by roadmapping different features. Identifying the MVP is crucial to delivering a focused, high-impact product that excels in its core function. By aligning our development strategy with clear business metrics, we can better measure success and make informed decisions on future iterations.

●  Next Project: Blend  ->

Connect.

Linkedin

Instagram

Contact.

peterlin.atelier@gmail.com

646-203-3270