14.03.2017 15:50

To combat bias at work, the design firm Doberman designed an app that charts gender bias through voice recognition.

Studies have shown that women face bias when they speak up in the workplace. Regardless of whether it's conscious or unconscious, that bias has real consequences. For instance, in a study in which Yale psychologist Victoria L. Brescoll asked male and female employees to evaluate executive performances, she found that female executives who spoke frequently were given 14% lower ratings of competence. Their chatty male peers, meanwhile, were rewarded with 10% higher ratings.

It's no wonder, then, why women on average speak less than men in meetings: These risks aren't just perceived, they're proven. The Stockholm-based design firm Doberman believes the first step toward erasing gender bias in the workplace is making it known. For that, it designed an app called GenderEQ that monitors and evaluates meetings based on voice recognition, then analyzes the data to show the percentage of time taken up by male and female speakers.

The app works on a phone or tablet, the idea being that the device can be set down on a table in the middle of a meeting and chart, in real time, who is speaking and for how long. The team at Doberman developed an algorithm that extracts data from the sounds picked up by the device's microphone. The app doesn't record the conversation or analyze the content of a person's speech—it won't infringe on the user's privacy—it just recognizes the frequency of voices and labels them as male or female.


Meanwhile, the app's interface illustrates whether a man or woman is speaking in real time, thanks to a line that looks like a sound wave fluctuating either on the side with a male symbol or the side with the female symbol (or both, if multiple people are taking at once). After the user ends the session with a push of a button, the line snakes vertically down the screen, becoming a line graph that shows an overview of speaking time during the meeting. It's exceedingly simple to use, with a tight and elegant design.

The app does bring up a few obvious questions, though: Does grouping voices into male and female voice based on the frequency and tone reinforce certain stereotypes, even as the app seeks to do away with others? And what about people who don't identify with either symbol on the screen? "We are very aware that gender identity is a much more complex subject than this," says Lars Ericsson, head of technology at Doberman. "We don’t think that this will solve the question around gender identity or gender equality. That’s definitely an important point to make. What we hope to achieve is to raise the awareness and fuel the discussion around how we interact and behave."


While the designers at Doberman know merely charting and displaying data on who talks the most in meetings won't solve gender inequality, they are hoping to spark a conversation about it. Users can share the data with their team, and chart how meetings evolve over time. The design of the app is simple, unobtrusive, and readable at a glance; the designers hope that being able to seeing the data in real time during meetings will make people conscious of their own behavior—whether that means speaking up more, or checking their judgment when others do.

"The meeting room is a big part of [people's] everyday culture at work, and it's a place where a lot of decisions are made," says Ericsson. "Having [the app] present in the room will raise awareness in real time to affect what's going on in the moment."

Originally published FastCodesign

Join the conversation! We'd love to hear your thoughts.

Please complete all fields below
Your name *
Email address *
Message *