Hear the line between two English vowels, from many voices, until your ear believes it. Vowel Border is ear training for learners who know the rule for ship and sheep but still hear the same word: minimal pairs spoken by the English voices on your phone, then a quick quiz, with every border scored on a map of confusable sounds. No account, no server, works offline.
What it does
- A map of 21 borders between similar English vowels, grouped the way learners confuse them: ɪ / iː, æ / e, ɔː / ʌ, uː / ʊ, diphthongs, r-colored vowels and triphthongs. Each border is colored by your score.
- Train: five word pairs per border (ship | sheep, bad | bed…), side by side. Tap a word to hear it, or Play all to hear the pairs alternate through several voices. Speed 0.8 to 1.0.
- Quiz: ten trials per round. AB mode plays one word and you pick which it was; ABX mode plays two references and a third word, and you pick which one it matched. Every answer shows which word it was, then the next one plays.
- Many voices, same word: the app rotates through the English voices installed on your phone (US, UK, Australian and more) so the contrast survives a change of speaker. It never asks a contrast that the chosen voice itself does not make.
- Honest results: each round shows the score, the border's score before and after, and the words you missed with the voice that spoke them. No streaks, no points.
- History by day and by border, an optional daily reminder, and everything stored on your phone.
How to use
- Open Map and tap the Train button on the suggested border, or pick any border in the list.
- In Train, tap words to hear them, or tap Play all. Listen for what changes between the left and the right column.
- Tap Quiz. A voice says one word; tap the word you heard. Each answer shows which word it was, then the next one plays.
- Read the result, then Again or Back to map. When a border passes 70 %, Try ABX appears for the harder mode.
- Open Voices to turn voices on or off, preview them, and see how to install more on your phone.
FAQ
Why several voices? One speaker teaches you that speaker, not the vowel. Research on phonetic training shows that hearing a contrast from many talkers, with feedback, is what makes it stick and carry over to new speakers.
Where do the voices come from? From your phone. On iPhone these are Apple's text-to-speech voices (Settings → Accessibility → Spoken Content → Voices → English lets you install enhanced ones). On Android they come from your text-to-speech engine (the Voices screen has a button to its settings). The app does not download voices or send any text anywhere.
A word sounds odd, or two words sound the same. Text-to-speech voices differ. If one voice blurs a pair, turn it off in Voices; the quiz already skips voices whose accent does not make a given contrast (for example ɔɚ / ʊɚ with British voices).
Why no microphone or pronunciation scoring? Vowel Border trains the ear first. Accurate production needs accurate perception, and a scored microphone test tends to grade the recording, not your hearing. A speaking mode may come later.
Does it need internet or an account? No. Everything runs on your phone and works in airplane mode. There is no sign-up, no ads and no tracking.