The Prompt
Come up with a new and innovative way that users either experience, consume, or create entertainment/media in less than 24 hours. Categories included Top 3 for Overall Idea, a Best Second Screen App category, and a Best Marketing Concept category.
The Inspiration
A company called KlugTV had created a second screen app that made it easy for user to learn about characters on TV and purchase products seen on any given show in real time as they are watching that program. It was an innovative way that enhanced e-commerce for stores and products affiliated with the show or episode, as well as enhanced how users discovered and purchased products.
This concept primarily benefits businesses as it helps them creatively market to consumers in a more seamless and effortless way. Imagine if this same technology could be used to benefit viewers consuming politics and various news outlets.
Below is an example of the Klug TV interface.
This prompts the question:
“How can second screen apps change the current news landscape and how will it influence this current climate of “fake news”?”
“Fake News” has always been around, but has become an even bigger issue since the previous election.
How can people stay on top of legitimate news and how can they inform themselves of both political figures and topics in real time as they are watching or listening to debates, news segments, and other news media coverage?
The User
Our user is someone described as “the uninformed who wishes to be informed.” This is the person, who, similar to those who create new year’s resolutions, they’d like to follow through, but the resolution or in the case, the choice to become more informed about political events and the world around them — simply falls off their priority list.
Life happens, and they simply forget. And as hard as they try, it is hard for this person to create a new routine in their already busy schedule.
Current Search Methods
TO BRING TRUE VALUE, WE HAD TO PINPOINT THE INTEGRAL DIFFERENCE AND BENEFIT BETWEEN LEARNING THROUGH UREKA VS. HOW PEOPLE CURRENTLY SEARCH TOPICS
One of the biggest pain points we hypothesized is a person’s ability to accurately remember what they want to search, as well as the memory to conduct the search. Things that can affect a person’s recall accuracy:
- Clarity of audio they are listening to
- Time it takes to open and load phone browser and enter their search query
- Rather than searching now, they’d like to search later but forget
These methods create large room for errors and blind spots.
Our Focus
TO TACKLE THIS, WE DETERMINED KEY NEEDS:
- The ability to quickly search
- Ability to integrate audio-recognition technology
- Determine what API’s we could pull from
The Solution: UREKA
BRIDGING THE GAP AND CREATING COMMON AND ACCURATE KNOWLEDGE
UREKA is a second screen app intended to bridge people of all levels of understanding and education in real-time through speech recognition.
With U-Reka, users can tune into political debates, broadcast news, or even an infomercial for a new marketed medicine and use their second screen app to identify the audio and pull key-words, information, and recent articles pertaining to the content they are watching
Our Focus
TO TACKLE THIS, WE DETERMINED KEY NEEDS:
- The ability to quickly search
- Ability to integrate audio-recognition technology
- Determine what API’s we could pull from
Current Search Methods
To bring true value, we had to pinpoint the integral difference and benefit between learning through UREKA vs. how people currently search topics.
One of the biggest pain points we hypothesized is a person’s ability to accurately remember what they want to search, as well as the memory to conduct the search. Things that can affect a person’s recall accuracy:
- Clarity of audio they are listening to
- Time it takes to open and load phone browser and enter their search query
- Rather than searching now, they’d like to search later but forget
These methods create large room for errors and blind spots.
Affordances
Through this app the user has the affordance of creating quality searches and removing room for error based on current methods of search that simply relies on memory, clarity of audio, and the persons ability to type quickly should they want to search something they’re watching.
- Information automatically retrieved based on audio (think something along the lines of Shazaam)
- Ability to identify speakers, get bios to help assess credibility and background, ie polticians, news anchors, etc
- Retrieve recent articles pertaining to topic
- Create visual text of audio with highlighted keywords and names for the user to click and conduct a deep-dive discovery
The Scenario
Our initial goal was to focus on a scenario where our user is just tuning into a debate. It is the third debate in a series of four presidential debates, Trump is talking about DACA and our user wants to learn more about DACA. In this scenario the points of interaction we witness are:
- The user initiating speech recognition
- The app identifying the speaker and outputting text from the audio
- The user clicking on a key-word highlighted within the text from the app
- The user moving to a dedicated info page regarding DACA
How It Works
Ureka will then pull up relevant articles and relevant hashtags. The point of this is to also speed up the search via motion. User can swipe right on a hashtag to conduct a deep dive search, swipe down to save for later, and swipe left to discard.
Process
CREATING A WORKING PRODUCT IN LESS THAN 24 HRS.
The process in the duration of this hackathon was unique. We had less than 24 hours to onboard our teammates, figure out our technical capabilities and which scenario and user flow we wanted to focus on.
First I pitched the idea to my team and made sure we were all on aligned with the same vision. My main concern as project lead/project manager was maintaining this aligned focus.
Because there was such a short amount of time, we needed to create timely check-ins with both the design team and the developers. The design needed to be compatible with the technical capabilities that are being discovered as they are being discovered.
Design Studio
Sketches, Medium, High-Fidelity
Technical Capabilities
On our team, we were lucky to have two front-end developers and one back-end developer. We wanted to utilize speech recognition as well as the capability to auto-retrieve information based on that soundbite.
Tech we implemented includes:
- Voice Recognition Google Cloud Speech API: Node.js Client
- Keyword Analytics https://cloud.google.com/natural-language/
- Metadata Crawler
