From Science Fiction to reality
I worked with the founding team at Waverly Labs to help launch their flagship product, Pilot Speech Translator. Taking machine translation to new heights, the team envisions a world where seamless, conversational translation is accessible to anyone, anywhere.
My role was to lead the design and development of all software, as well as head up the integration of the hardware, which took the form of a pair of specially tuned, noise cancelling earbuds. I assembled a team of designers and developers and set out to gather requirements and understand the constraints of the hardware, software and timeline. Throughout the process, I brought a user-centered perspective to the team of engineers and developers. Over the course of the project, my role evolved into that of a product architect: setting the vision, gathering requirements, and collaborating with the team of designers, developers and hardware engineers to ensure that features were executed to spec.
UX Design, Product Management
Tasks: Qualitative Research, User Interviews, Competitive Benchmarking
My first task was to build an understanding of our users. Building a product from a crowdfunding campaign has it’s benefits. The team had already assembled more than 20,000 backers, but had little understanding of exactly who they were and what they were expecting. I kicked off research with a survey of customer demographics and expectations for the product. Understanding basic demographics proved to be very useful in setting the tone for our user experience. From there I organized a series of 1 on 1 interviews, where I gathered invaluable information about how users planned to use the product, and in what scenarios.
User Experience Design
TASKS: USER FLOWS, WIREFRAMES, JOURNEY MAPPING, PROTOTYPING
Throughout the project I spent a lot of time working with stakeholders on our team to turn a lofty vision into reality. One of the biggest challenges was balancing our long term goals, and the vision of an ideal product against the limitations of today’s hardware, software and machine translation in general. I created flows and wireframes for many of the apps core functions, from registration to product setup to conversation.
This sketch is a concept for Pilot Earpiece owners and guest users entering a conversation. Though both users must install the app on their smartphone, we wanted to simplify the flow for guests as much as possible. The host creates the conversation and enters all crucial information to get the conversation started.
Much of our effort from a UX perspective was focused on the integration of hardware and software. I created many journey maps like this one, in an effort to coordinate all aspects of the experience in one place: hardware feedback, on-screen actions, and physical interaction between both users.
A more finished version of the journey map, this time with proper wireframes.
These sketches are an exploration of how the home screen could function for guest users and Pilot Owners alike.
User Testing & Refinement
Tasks: Prototyping, Usability Testing, Research Facilitation
This critical part of the process presented some unique challenges. With so many moving parts: multiple users (who don’t speak the same language), multiple phones, and a set of earpieces, orchestrating user tests required a good deal of planning. We prototyped this scenario by recruiting a handful of our backers, then pairing them with non-english speakers and observed as they used the product.
One of our usability tests included two users, one English speaker and one Spanish speaker, conducting a conversation together.
Tasks: User Interviews, Sprint Planning, Backlog Management, Quality Assurance
Working with a very young startup meant my role evolved over the course of the project. I was fortunate enough to work on many aspects of this product: collaborating with the hardware team, setting out a vision for the software, and working with our team of developers to design and execute an iOS and Android apps. My work is currently focused on managing backlog of tasks, prioritizing enhancements, and learning from users as we roll out the product.
Machine translation (in terms of consumer technology) is still in it's infancy. While we have lofty visions for the future of conversational translation, the first iteration of the product has to allow flexibility, and communicate exactly what’s being translated so that users can build confidence and react in the event of mistranslation. Much of our effort was spent dreaming up an ideal scenario, then scaling back to make the most of the technology as it exists today.
Designing a multi-user experience with custom hardware has it's challenges. For as much testing as we did with design prototypes, it wasn’t until we had a rough working version of the app that we could truly understand how both users would interact with it, and each other. We found it most useful to educate the conversation host, and let them handle setting up the conversation, since guests are likely first-time users.