Giving Science a Voice: a UX Story on Utilizing Voice Data
In the assignment, I was tasked with constructing a lean solution phone app elevating the use of voice data. While voice data is a new technology, it has become integrated into our lives with Siri and Alexa, finding new ways to dispense information to us. So I tried to find inspiration within my own life to solve this problem. Currently, I live in Houston, home to the largest medical center in the world. My best friend also just happened to have recently graduated with her Ph.D. in Data Science. After a brainstorming session, I decided to tackle the field of scientific research. How could I use User Experience Design to optimize voice data in a way that could benefit scientific research?
I walked into this project with a few assumptions: the technology we are already familiar with is not tailored for scientific research. Have you ever tried to ask Siri for help with a medical term only for her to botch it? Even talk-to-text translations have a difficult time with scientific jargon. Several of the terms used in science are not phonetically written so even when you speak slowly, the AI does not compute.
RESEARCH
I interviewed 5 people for this project: 3 Ph.D. Scientists, 1 Lab Manager, and 1 Ph.D. researcher. I wanted to understand their frustrations in the lab, where the pain points were when analyzing data, and how we could aid in resolving user error with voice-activated assistance.
Interview Questions
- What is your current role in the scientific field?
- How long have you held this position?
- How long have you been performing science experiments?
- Walk me through a typical experiment
- What are your frustrations when performing/analyzing an experiment?
- How often do you find yourself taking off your gloves to set a timer?
- What would make your experiments easier?
- What are your frustrations with the data?
- Do you have any other insights you’d like to share?
Notable Quotes:
- “My greatest frustration is not having a unified system where everything is, I just want to print out my protocols and get to work” -Scientist
- “I hate when I don’t plan well and have to take off my gloves to use a calculator or set a timer” -Scientist
- “Everything needs to be recorded and it’s very easy to lose track of things” -Lab manager
- “I just want a list of experiments done on the same sample instead of hunting down the scientist who performed the experiment” -Researcher
Synthesis
Lab Scientists perform experiments frequently and need to catalog the data of the materials used, the experiment, and the protocols used. Much of this is repetitive but becomes exhausting to catalog and can result in missing data points.
Labs will also perform several experiments on the same sample but make it difficult to search all the variations that were performed. To relieve this some labs have a protocol to barcode their experiments to reduce user error and keep records, but this is not always enough. Barcode data is not always easy to catalog since the stickers might be too small or initially not enough information was given.
One interesting thing I did find out was how irritating it is for scientists to constantly be changing gloves. Forgetting to do a calculation or having to set a timer means removing your current gloves, washing your hands, completing an activity, washing your hands again, then putting on a new set of gloves. Nothing is worse than having to try and shove moist hands into a pair of latex gloves. Not to mention that over time latex allergies occur, if you don’t already have a skin condition. This made the voice-data application so much more necessary.
FLOWS + SKETCHES
Initially, I only had two user assignments but soon realized that lab managers also have inquiries that need to be addressed without hunting down the researcher. I also ended up changing how the user would get into their workflow, creating a shared database that you can add collaborators to.
Ultimately, once the experiment and protocol are set, the user can use the voice prompt to print, set alarms, stop alarms, and use a searchable talk-to-text notebook during their experiments.
FINAL THOUGHTS
Ultimately, I think this would work better as a browser-based application. Phone screens are limiting and difficult to consume large amounts of data on. I would also like this to become a platform where researchers can collaborate and peer-review data, furthering scientific research.
Other functions I would like to include would be to integrate R-markdown as an export function and also create a voice-activated device capable of advanced math to alleviate the hassle of calculators during an experiment.
In the future, I would like to expand my data pool and interview more researchers and lab managers. I believe they have valuable insight into what makes for strong data collection and how to keep it organized. Ideally, I would have also liked to conduct a contextual inquiry study to see first-hand how scientists perform their experiments and to better understand when and where it would be appropriate to have voice-activated assistance.
