FreeSpeech

U.S. Citizenship & Immigration Services

Asylee Interview Prototype

System Design Prototype, Summer 2019
 

Team Members: Johncarlo Cerna, Isaac Robinson, Noah Houpt, Amit Rajesh, Sarah Frost (Software Engineering Fellows)

Role: Lead Designer
Focal Points: Building Trust + Accessibility

As part of the Civic Digital Fellowship, Coding it Forward hosted a hack-a-thon geared towards social good. Splitting into groups based on interests, my team and I decided to focus on the current inefficiencies of the interview process that asylum-seeking refugees go through from both the perspective of the interviewer and asylee.

 

Given three hours, our intervention and prototype(s) definitely aren't perfect but they represent a concept and conversation that has the potential to move forward with more time and iteration.


 

 

 

(Left to Right: Noah Houpt, Amit Rajesh, Dorcas Lin, Isaac Robinson, Johncarlo Cerna)

 

In order to understand the current USCIS system, we mapped out the user journey and major points of interactions to find weaknesses.

The biggest pain point was found in the manual transcriptions and editing that has to be done by the adjudicators while interviewing the asylee. It's often not verbatim and only upwards of five interviews are completed a day per interviewer. 

After transcription, the text file goes into a system called Global and becomes a legal document with the ability to rephrase what the asylee has said (for purposes of context).

As the designer, I led the discussion for changes made in the front-facing interface in order to better inform the accessibility of the current interactive experience.

Research

Currently, Refugee, Asylum, and International Operations Directorate (RAIO) adjudicators conduct interviews with asylum and refugee seekers. These interviews are ~90 minutes long and require the adjudicator to ask questions while transcribing conversations as accurately as possible.


 

 

Intervention Speech-to-Text Transcription Service

Wanting to emphasize human interaction in a naturally detached environment where adjudicators are currently more focused on the computer screen than those relaying the information, our intervention utilizes automated transcription to reduce stress and increase efficiency for adjudicators. For the asylee, this involves a separate front-facing UI with confirmation steps throughout the interview in order to establish comfort and trust.

Features

  • Local Speech-to-Text API

  • Differentiates between speakers 

  • Edit and comment/note support on transcript 

  • Secure: deletes all audio recordings by the end of the interview

  • Uses existing tech in asylum interview rooms


 

 

Front-facing iPad UI for Asylee

 

Design Choices Reactive Recording UI

Leading the team, I mapped out a persona and its user journey based on the pain points we indicated before. At this point, our main priorities were:

1. Building Trust

2. Reducing Stress

3. Increasing Efficiency

Following these three priorities consistently throughout the design process allowed us to design for inclusivity and accessibility. Another major component we wanted to emphasize with our design was to analyze and emphasize what the fundamentals of interactions were for cohesivity in the user experience.

It was important for us to combine the microphone icon with the changes in color for the sake of redundancy and understanding that the user is being recording. This further equalizes the experience for those who might have low reading proficiencies.

Design Choices Info Deletion Pop-Up

One of our main priorities of building trust was what led to the confirmation of information being deleted. We wanted to ensure that the asylee understood what was happening every step of the way without skepticism. The pop-up box is a component of added security and the physical push of the "okay" button solidifies the process.

Reflection

This was an interesting problem to tackle with heavy time constraints placed upon us. As a designer, it was difficult to put myself in the shoes of the core stakeholders to truly design for access through interaction. At the end of the day, accessibility isn't prioritizing any one audience group over another, rather, it's meant to equalize and create the same experience that benefits all.

The priorities and fundamentals kept the team on track and on the same page even when a some of us were working on the front end for the asylee, others on the front end for the adjudicator, and more on the backend for the speech-to-text translator.

 

Working within an interdisciplinary team isn't common for me in design school, so I was excited to be able to garner each person's perspective and lived experiences to create something that tackles our society's most prevalent issues firsthand.