Viewmo

Monitor viewer expressions and emotions!

Comment

Table Number: 325

Inspiration

Our main inspirations for building an app were to create an app that can handle multiple use cases and because one of our team member's cousin (Ryan's) from Ho Chi Minh City, Vietnam is in need of seeking feedback from her television personality shows. As a television show host, she is always in need of feedback, especially user engagement and emotional feedback, in order to improve her game shows.

What it does

Viewmo allows users to participate in allowing video recording and analysis of their facial expressions to record their emotions while watching their inputted video throughout the website. Providers are able to view the video URLs alongside the associated statistics regarding the general audience's emotional reactions.

How we built it

Our web application is mainly built upon Hume AI technology. Viewmo is a full-stack application that employ React (Next.js) which utilized Server Side Rendering and App Router. Our backend infrastructure consists of FastAPI for a reliable and easy to set up communication between the client and the server. We also use WebSockets to maintain a consistent state and handle data stream of images capture in the frontend.

From the backend, we process the data by using the Expression Measurements API by Hume to extract out emotions in real time throughout the duration of recording. Then, we process the recorded emotions as an array and reliably store it in MySQL database hosted on Amazon AWS server to keep track of the record of users' reactions when they consume a certain form of media. Lastly, we take advantage of AWS to host the database server to improve our workflow and productivity.

Challenges we ran into

There were some challenges that we encounter during the development of Viewmo. One of the challenges that we faced were due to our inexperience with the Hume technology, we were unable to make use of the EVI technology at Hume AI to integrate it in our app. Therefore, we had to make a crucial decision to switch to Expression Measurement API technology to measure users' emotions when they consume any form of media.

Accomplishments that we're proud of

Some accomplishments we were proud of were being able to utilize Hume AI technology in our web app, especially we were unfamiliar with the tech, but was able to learn it in short amount of time. We were also unfamiliar with websocket communication protocols, so being able to learn websockets in such short amount of time is quite impressive which fully enable us to fully take advantage of them to implement them in our app.

What we learned

Some things we learned throughout and after developing our app was a lot of information about emotion AI and the importance it can play in applications. We also learned a little bit more about proper time management and great collaboration skills as putting emphasis into these skills pushed us to develop a well done app that is finished on time.

What's next for Viewmo

The future for Viewmo includes integrating ads to the videos, providing opportunities for corporations to advertise their products to more potential consumers. A possible feature we were highly thinking of adding to our app includes a in-app coupon rewards feature that would reward users coupons with discounts for the products advertised after viewing.

Share this project:

Updates