1. 程式人生 > >How iLingual Uses Computer Vision AI to Translate and Learn Languages

How iLingual Uses Computer Vision AI to Translate and Learn Languages

Omar Alexander Abbas and Raymona Singh are the creators of iLingual, a free app that enables users to learn languages by bringing together machine learning and language translation. We recently got to speak with Omar about his work and use of Clarifai. Check out the interview below!

We'd love to learn about you and Raymona! How did you get to where you are today?

Hi, I’m Omar! I was born and raised in NYC, and I'm a recent-ish graduate from Baruch College. I majored in Economics but found a knack for coding during my last semester. It honestly all started by wanting to understand what’s happening beneath the surface of all these things we use in our daily lives.

I started off learning Python through YouTube videos, and when I eventually wrote my first few lines of code that printed “Hello world,” I was hooked. After I had a few more languages and a decent amount of jargon under my belt, I started reaching out to more experienced developers, reading textbooks/articles, and practicing whenever I could. The world of coding is a constantly changing environment, so it’s always exciting to see what’s around the corner. I've created a few apps so far, but besides this I'm always open to learning new things relating to the fields of physics and philosophy.

Raymona, my partner-in-crime, moved to NYC a few years back and also graduated from Baruch with a major in Philosophy. She's working her way to law school and pretty much enjoys anything related to the tech field. She's the type of person you go to when you want to have a conversation about Robot ethics--that's the best way to sum it up.

What is the name of your application, and what does it do?

iLingual is a modern way to learn a new language. We incorporated artificial intelligence like photo and voice recognition so users find it more convenient to access real-time language translation. The process is simple, you either select a photo from your library, snap a photo, or use your camera for real-time detection. The AI predicts what it sees and gives the user their desired language translation in both text and speech.

Why did you build this app, and what inspired you? Tell us about your “Eureka!” moment.

We were basically testing out photo recognition for the fun of it and the idea of language translation developed as we went along. We started promoting it on various social media platforms and people seem to be really interested in it, so we are constantly improving it.

Love that you're constantly improving the app. What tweaks have you made since the launch?

A bunch! In terms of behind the scenes tweaks, we’ve worked on reducing the app’s energy consumption to avoid straining battery life. We’ve also moved a lot of the network related processes to our end. So in the situation a user is on a limited data plan, they won’t have to worry. Another big one was the size of the app. We didn’t want to be the first to the gauntlet when the “Storage almost full” alerts appear. On the user-facing side, we’ve worked to keep everything modern and user-friendly. We looked for ways to integrate the recurring design/functionality trends that other apps were adopting.

How did you build it? What programming language did you use? What about other tools?

The app is built and maintained using Xcode and Swift. Behind the scenes, we have Firebase as our back end written all in Node.js.

What was the best part about working with the Clarifai API? Did you compare the Clarifai API to other image recognition services? If so, which ones?

One of the first features we wanted to implement was image recognition language learning. It was our first toe dip into the world of AI, so we didn't know what to expect. We shopped around but found that most APIs were lacking in something, IBM Watson was offline and free but too large, Google was relatively complex but pricey, and like Goldilocks in the middle bed, Clarifai was there. Not to get all infomercial, but it was a breeze setting up. Also, iLingual would result in a high amount of recognition usage, and Clarifai offered flexibility as far as pricing, so that's always a plus.

Anything else you’d like us to know about or share?

We're going to be releasing a speech to text update soon, it'll allow you to create a transcript in almost 60+ languages of practically any video you find online. That includes Instagram, Twitter, YouTube Facebook, etc. After that, there will be a museum update, so be sure to stay in touch!

Thanks for sharing, Omar!

To try out the app, download it for free here.

Want to connect with the iLingual team? Follow along the app’s adventures on Twitter and/or Instagram and don’t forget to give some props to Omar + Raymona.

To learn more, check out our documentation and sign-up for a free Clarifai account to start using our API – all it takes is a few lines of code to get up and running! We’re super excited to share all the cool things built by our developer community, so don’t forget to tweet @Clarifai to show us your apps.

New call-to-action