POLLYLINGO

A hackathon video conference translator project. Pollylingo gives users the ability to speak in their native language. All audio will be translated to the recipient's choice of language. For example, if a Spanish speaker is talking to a Russian speaker, the Spanish speaker will speak in Spanish and the Russian speaker will hear Russian. This mini-project was built with React, Node.js, Socket.io, Web RTC, AWS.


See it in action: Linkedin

MY ROLE
UX, Research, Web Design/Development

zz-2

 

Overview

I helped design and develop this hackathon project with two engineers, one designer, and one data scientist. 

The goal of this project was to help remove the language barrier during multilingual conference calls and create a seamless experience for the user.

 

 

Conceptualization 

a1
p3

This logo was selected because we wanted to convey to the user quickly that the main function of this app is translation.

 

Wireframe

  • Landing Page - Our goal was to provide a simple experience without clutter so the user can get started right away. The elements are placed at the center of the screen. The user's natural eye movement guides the user from top to bottom. 

  • V1 - We had a large button on the bottom of the screen that said "Press and hold space bar to record". This was the biggest element on the screen and was colored purple to draw attention to the element. This was intentional so users knew what the next step was. There was no sound during the call until the user spoke while holding down the button.

  • V2 - After we built the hands-free feature that allowed users to translate anything that was said while the mic was on, we removed the large button and added a mute button. 

 

Group-13
revision

For a better user experience, I would place the transcript at the top middle section of the video screen so that the eyes will be closer to the camera. This will provide a better eye level during the call instead of looking down.

The mute and disconnect button would be at the bottom middle section. We are conditioned to search for controls on the bottom of the screen from other video conference software like Zoom, Skype, and Gotomeeting. During video conferences people who are not speaking typically mute themselves. Placing it at the bottom middle is intuitive for the user.  

mm

 

Conclusion

To take the project further we could add more personalization to the call. A voice model can be created for each user to reduce background noise interference and allow the syntenic voice playback to sound like the original voice.

 

Let's Connect

Feel free to reach out for any questions, collabs, or to say hello ?

Dribbble@2x
Linkedin@2x