English Shadowing open the app →

English Consonant Sounds That Change Words

Vowels get most of the attention. I wrote about them in English vowel sounds, and they are easier to hear. But English consonants do more of the work when a word has to be identified. A small change in one consonant turns light into right, sin into sing, or back into bag.

Most learners can hear vowels. They often miss consonants because their own language uses a different set. The useful part: consonants are visible. You can see your lips, teeth, and tongue. That means you can correct them with a mirror and a slow video.

The five sounds that matter most

These five cover the majority of consonant problems across learner languages. You do not need to be perfect on all English consonants. You need to control these.

/θ/ and /ð/: think and this

Put your tongue tip between your teeth. For /θ/ in think, push air out. For /ð/ in this, do the same but add voice. You should see your tongue in a mirror. If you replace these with /s/, /z/, /t/, or /d/, listeners usually understand from context. But if it happens in every word, your speech sounds fuzzy and takes more effort. Practice words: think, thick, bath, this, that, breathe.

/r/ and /l/: light and right

These are not one sound. For English /r/, the tongue pulls back and curls slightly; it does not touch the roof of the mouth. For /l/, the tongue tip touches the ridge just behind the top teeth. Light and right are different words. So are play and pray, glass and grass. This is a high-cost error because it changes meaning immediately. Practice slowly: say la, la, la and feel the tongue tip tap. Then say ra, ra, ra and feel the tongue pull back.

/v/ and /w/: vine and wine

For /v/, put your top teeth on your bottom lip and switch your voice on. For /w/, round your lips and do not let your teeth touch anything. The pair vine and wine is the classic test. Learners from Spanish and Russian backgrounds often mix these. Saying very well is a good practice phrase because it switches from /v/ to /w/ quickly.

/ŋ/: sing without a hard g

The sound at the end of sing is /ŋ/. The back of your tongue touches the soft palate, and the air goes out through your nose. Many learners add a hard /g/: sing-guh. That makes sing sound like sink with an extra vowel. Sin, sing, and sink are three different words. The first ends with the front tongue on the ridge, the second with the back tongue raised, and the third adds a stopped /k/. Try all three in order: sin, sing, sink.

Final voiced and voiceless consonants: bed and bet

English uses voice at the end of a word to separate pairs like bed and bet, bag and back, have and half. The mouth position is almost the same. The difference is whether your vocal folds keep moving. Hold the final consonant for a full second: bedddd, bagggg. Then shorten it. If you stop the voice too early, bed sounds like bet. If you add a vowel after a final consonant, bed-uh, you protect the voice but you have added a sound that English does not want.

Mouth positions at a glance

Sound Mouth position Pair Cost if wrong
/θ/ /ð/ Tongue between teeth; /θ/ air only, /ð/ voice think / this Low in context, but frequent substitution makes speech unclear
/r/ Tongue curled back, no roof contact light / right High: word identity changes
/l/ Tongue tip touches behind top teeth light / right High
/v/ vs /w/ /v/ teeth on lip, /w/ rounded lips vine / wine Medium: some pairs change
/ŋ/ Back of tongue up, nose air sin / sing / sink Medium: adding /g/ sounds unnatural
Final voiced vs voiceless Voice on vs voice off at the end bed / bet, bag / back Medium-high: tense and meaning change

Consonant clusters: no extra vowels

English packs consonants together. Street starts with three consonants: /s/, /t/, /r/. Many learners add a vowel before the cluster and say estreet. The word tests ends with the cluster /sts/. That is hard, but the answer is not test-uh. Adding vowels between consonants changes the rhythm and can create new words. Say sport, not suh-port.

Say clusters slowly first. Break them into parts: sss-t-r, then blend. Use the mirror to check that your mouth moves once through the cluster. Then speed up to a normal sentence.

Which mistakes actually cost you

Not all consonant errors are equal. Aspiration is the puff of air after /p/, /t/, and /k/ in words like pin and tin. If you say these without the puff, you will have a slight accent, but no one hears a different word. The /θ/ and /ð/ sounds are often replaced with /s/, /z/, /t/, or /d/. In context, people guess the word. But if you also drop final consonants or mix /r/ and /l/, the listener has to work harder. An accent is fine. A word error is not.

Two or three errors in one sentence can make I rent a flat sound closer to I lent a flath. The grammar may still work, but the words are no longer exact.

Practice: mirror first, then shadowing

Start with minimal pairs. The page minimal pairs practice gives you word pairs like light/right and bed/bet. Say them while watching your mouth in a mirror. Do the same pair ten times, then switch. Then put the words into short sentences: The right light is on the left. That sentence forces you to move between /r/ and /l/.

After minimal pairs, move to shadowing. Shadowing means speaking along with a video while you listen, at the same time or just behind. The longer explanation and different variants are on the what is English shadowing page. For consonant work, choose a video where the speaker is clear and not too fast. Slow it to 0.75x or 0.5x in the English Shadowing app. You can paste any YouTube link or search inside the app. Under every spoken word there is a translation into your own language. The currently spoken word is highlighted as the video plays.

Listen for the consonant you are working on. Then repeat the sentence with the same mouth shapes. Tap a sentence to loop it. If you tap a word, the app saves it to a vocabulary list with spaced repetition flashcards. That is useful for words like vine and wine, where you need repeated checks. The full sequence is described in improve English pronunciation.

This video is built for repeating and works well for consonant shadowing.