Minimal Pairs Practice: Train Your Ear Before Your Mouth

Pronunciation 8 min read

Headphones and blank flashcards laid out for listening practice
Listen for a week before you speak. The ear leads and the mouth follows.

If you have practised a sound for months without improvement, the problem is almost certainly perception, not production. Your brain categorises the English sound as the nearest sound in your own language, and once it has been filed there, no amount of repetition changes it.

Minimal pairs — two words differing by exactly one sound — are the standard tool for reopening that category, and they are simple enough to do on a commute.

Why the ear comes first#

Infants can distinguish every sound contrast in every language. By about a year old they have specialised in the contrasts their own language uses and stopped hearing the others as different. That specialisation is efficient and it is why adults hear a foreign contrast as one sound.

  • If ship and sheep sound identical to you, you will produce one word for both.
  • No amount of mouth practice fixes this, because you cannot check your own output.
  • Perception training reopens the contrast, usually within one to three weeks.
  • Production then improves quickly, sometimes without direct practice at all.

This is the reason a learner can say “I have practised this sound a thousand times and it has not changed”. The practice was aimed at the wrong system.

The pairs worth your time#

Minimal Pairs Practice: Train Your Ear Before Your Mouth — The pairs worth your time
ContrastExample pairCommon for speakers of
Short i / long eeship – sheepSpanish, Italian, Turkish, Arabic, Hindi
Short u / long oofull – foolMany, widely
b / vberry – verySpanish, Korean
v / wvest – westHindi, German, Dutch, Polish
l / rlight – rightJapanese, Korean, some Chinese varieties
Open a / mid ebad – bedFrench, Portuguese, Turkish

Pick the two or three that match your first language. Working through the full list is a way to make no progress on any of them.

A ten-minute daily routine#

  1. Listen only. Play both words of one pair, ten times, without repeating them aloud. Just listen.
  2. Discriminate. Have someone (or an app, or a shuffled playlist) play one at random. Guess which. Aim for 90 per cent before moving on.
  3. Then produce. Say both words, record, and play back the next day so you hear them without the memory of intending them.
  4. Put them in sentences: “The ship is cheap” and “The sheep is cheap”.
  5. Move to the next pair only when the first is automatic.

Do not skip step one. Learners consistently rush to production and then wonder why the sound has not changed after a month.

Making your own material#

You do not need an app. Any text-to-speech tool, or a dictionary with audio, will read both words of a pair. Record twenty randomised repetitions into a single file and use it as your discrimination test. The one thing to avoid is testing yourself with your own voice — you will hear what you intended rather than what you produced.

When minimal pairs are the wrong tool#

They are for sound categories, and a great many pronunciation problems are not about categories at all.

  • Rhythm and stress problems need clapping and shadowing, not minimal pairs.
  • Consonant cluster difficulties need slow build-up, adding one sound at a time.
  • Intonation needs whole sentences and imitation of a model.
  • Confidence and volume are not pronunciation problems at all, though they are often mistaken for them.

If listeners understand your words but find you hard to follow, the issue is rhythm. Minimal pairs will not help and will consume the time that would have.

Frequently asked questions

What are minimal pairs in English?

Two words that differ by exactly one sound, such as ship and sheep or light and right. Because everything else is held constant, they isolate a single contrast, which makes them the standard tool for training both perception and production of a specific sound.

How long does it take to hear a new sound contrast?

With daily discrimination practice, most learners reach reliable perception of a single contrast in one to three weeks. Production usually follows within a few weeks after that, and often improves partly on its own once the ear can check the output.

Do I need an app for minimal pairs practice?

No. Any dictionary with audio or a text-to-speech tool will produce the pair, and a shuffled recording of twenty repetitions is a perfectly good discrimination test. The essential element is that the test voice is not yours, because you cannot judge your own production reliably.

minimal pairs practiceminimal pairs englishship sheep pronunciationenglish sound discriminationhear english soundspronunciation training exercises

All guides

Last updated 2026-08-06 by spokenenglish.info · About us

Written by teachers

Every guide is researched and written by our editorial team, not spun from other sites.

Reviewed on a schedule

Each guide carries the date of its last review, and we publish the date even when nothing changed.

No paid placements

No school, app or course can buy a mention, a ranking or a link here.

Twelve languages

Every guide is translated, not machine-popped — each language has its own URL and its own review date.

Free to read

No email wall, no locked lessons, no trial that turns into a subscription.