EPHEMERAL
type
Team project
duration
October–December 2023
role
research, user flow, UI/UX design, wireframe, user persona
Bring your Dream to Life.
Dreams are usually hazy and often hard to remember once you wake up. We wanted to create a product that could help users recollect their dreams with the help of AI.
Ephemeral seamlessly integrates into users' daily routines, allowing them to curate and explore dreams upon waking. Users can choose dream moods, generate dream images, and delve deeper into their dreams with an AI-powered analysis tool. Simultaneously, Ephemeral fosters a vibrant dream community, contributing to users' overall well-being through a sleeping aid feature for a restful night's sleep.
inspiration
Dreams are where our thoughts and imagination come to life, but they often fade as soon as we wake up. As statistics show, few people can remember their dreams clearly. Our project aims to address this issue.
CBS, 2021
“Only 1 in 10 adults in America responded they always remember their dreams. Most said they sometimes remember their dreams.”
problem statement
During these dream-filled moments, we find ourselves immersed in a unique realm, where the boundaries of reality blur and our experiences unfold in a surreal landscape known only to us.
How can we help people remember their dreams more clearly and share their experiences with others?
research
We decided that the best way to tackle this problem was integration of AI tool in the product. To make dream that is imaginative, blurry, and momentary, AI image generation was a perfect tool to help visualize a dream within a moment. To find the best AI tool for our dream visualization, we tested out various existing AI text-to-image generators.
solution
Based on our research and testing, we decided that Stable Diffusion would be the best AI model for the product.
01
Most accurate results based on the text that reflected the surreal feeling and creativity of the dreams.
02
Lightweight architecture so that it can create high-quality images within seconds.
03
Additional filters for more accurate visualization, including negative prompts (choose what to exclude) style selection, and image reference.
market research
To find out opportunities and gap in the existing market, we analyzed competitor’s key features and availability, as well as the user reviews.
opportunity
We found that there was no app that had a sharing feature with default visualization and sleeping aid provided. Present apps fell under one or the other categories and not in both. So, we used these insights in our project and created an app that provided all these features.
target user
Our target users are individuals aged 12 and up who want to track and enhance their dreaming experience. The app’s main feature, dream visualization, relies on detailed dream descriptions for optimal results.
Age 12+ who wants to keep track of their dreams as well as improve their dreaming experience.
persona
Creating a persona who represents our key user was helpful to empathize with the users and further understand their needs.
empathy map
We created an empathy map after speaking with 5 participants to further illustrate our user’s needs, motivations, and experience.
user flow
This diagram shows a clear roadmap of the user’s journey within our application, covering every possible path.
information architecture
The insights from user flow exercise helped us solidify our mobile app structure.
low fidelity wireframe
We conducted user testing with low fidelity wireframe to find rooms for improvement.
mid fidelity wireframe
After the first feedback, we conducted second round of user testing with mid fidelity prototype.