When learning a language, revision systems are everywhere. Flashcards, spaced repetition apps, vocabulary trainers. Yet as a learner I kept running into the same frustration. I had to work through a long list of things I already knew before reaching the few items that actually required attention. The experience often felt inefficient. I found myself scrolling through obvious answers simply to unlock the next difficult item.

This led me to rethink how revision could work. Physical flashcards behave differently from most digital systems. When you revise with a deck of cards, the difficult ones naturally come back more often while the easy ones slowly drift away from your focus. You are not following a calendar schedule. You are responding to difficulty and attention. I wanted to recreate that experience digitally. The result was a revision module inspired by the Leitner system but designed to behave more like a shuffled deck that continuously surfaces the items that still cause trouble.

To test whether this approach could deliver the kind of learning flow I was looking for, I needed a concrete learning context. Pronunciation training turned out to be a good fit. Before working in technology I used to teach English, and one thing always stood out to me. In schools there is a strong focus on exams, grammar, and written accuracy. Pronunciation often receives far less attention simply because there is limited time in the curriculum. Yet in international working environments, pronunciation plays a surprisingly important role. When English becomes the shared language between people from different countries, pronunciation shapes the first impression and can influence how easily communication flows. In a way it works like clothing. It is not everything, but it affects how the message is received.

Because of that, I decided to use British English minimal pairs as the learning content for the prototype. Minimal pairs are words that differ by only one sound, such as ship and sheep or full and fool. These contrasts are small but meaningful, and they are ideal for short, focused exercises. I designed two interaction modes. In one, the learner hears a sound and chooses the correct spelling. In the other, the learner sees a word and selects the correct pronunciation. Together these exercises train both listening discrimination and awareness of sound patterns in written English.

The project also gave me an opportunity to explore Flask, a Python framework,  in the context of a real product idea. Rather than learning the framework through isolated tutorials, I wanted to see how it behaves when supporting a small interactive learning tool. This made the technical exploration feel connected to a concrete design problem.

After building this early prototype I tested the app with a small group of learners and gathered feedback on the interaction and learning flow. The learning mode still has room for improvement, but the revision module itself produced encouraging results. Test users reported noticeably less friction compared to revision features they had previously used in popular learning platforms. I experimented with several of those systems during my research, although I prefer not to name specific companies since the goal was not to criticize existing products but to explore an alternative design approach.

PrawnApp is therefore less about building yet another pronunciation trainer and more about experimenting with a different model of revision. Pronunciation practice simply provided a meaningful context to test whether a deck based revision system can keep learners focused on what actually needs work. The early feedback suggests that this approach has potential, and the project will continue evolving as I refine both the learning content and the interaction design.

Curious about why I called it the PrawnApp? The visual style of the interface was partly inspired by a popular language learning platform that uses an owl as its mascot. As a small joke I chose a prawn instead. It felt fitting for an app about pronunciation, and the name plays on the sound of the word itself: prawn-unciation.