How to pronounce Silent T in Clusters t→∅ in American English

/t/ is dropped entirely — the surrounding consonants flow together without the T.

Start here

When /t/ gets sandwiched between two other consonants in American English, it goes silent — what linguists call cluster reduction. Exactly comes out as exacly (T between /k/ and /l/). Mostly becomes mosly. Softly turns to soffly. Three consonants in a row force the tongue to pause, and dropping the middle one keeps the sentence moving. The same shortcut crosses word boundaries: must be flattens to mus be, just looking to jus looking. One important exception: when the first consonant is an R, the T survives. Partly, heartless, and startling keep their T (often as a glottal stop or unreleased T) rather than dropping it.

How it triggers

Watch it happen in real phrases.

Three example phrases showing exactly when this rule fires.

exactly

The /t/ is sandwiched between /k/ and /l/ — three consonants the tongue has to articulate back-to-back. American speech drops the middle one, letting the sound glide directly from /k/ into /l/. Exacly. Same logic in directly (/k/ + /t/ + /l/ → drop the /t/), strictly, perfectly.

softly

Different cluster, same mechanic. The /t/ sits between /f/ and /l/. The tongue skips the /t/ and lets /f/ flow into /l/: soffly. Same in postpone (/s/ + /t/ + /p/ → /pos-pone/), investments, artist's.

must be

Cross-word version. Must alone keeps its /t/ because the cluster is just /st/ at the end. But put it before another consonant and the /t/ drops: must be becomes mus be, must go becomes mus go. The rule fires whenever the /t/ ends up sandwiched, whether the cluster forms inside a word or only emerges at a word boundary.

Where you'll hear it

In real American conversation.

Listen for words ending in -stly, -ctly, -ftlyexactly, mostly, directly, softly, strictly. The T almost never survives. Cross-word: last night, just fine, most people. News anchors, podcast hosts, sitcom dialogue — the silent T is the default everywhere. Pronouncing every consonant in the cluster sounds over-careful, like you're reading the word from a slide rather than saying it.

Underlying sounds

The sound this rule silences.

Click to explore how /t/ behaves across all the environments where it appears.

Hear the silent T in consonant clusters

Words where the T drops out of a cluster.

Tap any to hear the surrounding consonants flow together with no /t/ between them.

FAQ

Common questions about Silent T in Clusters.

Why do Americans drop the T in words like "exactly"?
Because pronouncing three consonants in a row breaks the rhythm. The /k/, /t/, /l/ in exactly forces the airflow to stop completely for the T before releasing into the L; dropping the middle T lets the sound glide directly from K into L. It's about flow. American speech weights smooth rhythm over hitting every letter, and dropping the T is what keeps the rhythm intact.
Is it sloppy or incorrect to drop the T in consonant clusters?
No, it's standard. Even in professional contexts, news broadcasts, and prepared speeches, native speakers drop /t/ when it sits between two other consonants. Pronouncing every letter in mostly or softly registers as careful or non-native to American ears. The dropped T is correct American pronunciation. It's how the rule works in connected speech, not a casual variant of it.
Does the dropped-T rule cross word boundaries?
Yes, as long as the T sits between two consonants. Must be becomes mus be; last night becomes las night; next door becomes nex door. The tongue skips the T and links the surrounding consonants directly. Force a crisp T at the end of just in the phrase just looking and you break the natural rhythm — the listener registers a tiny pause that wasn't supposed to be there.

Master the Silent T in Clusters with an AI coach.

SayWaader is the AI pronunciation coach for American English. Practice 5 minutes a day. Get a 5-axes accent assessment. Sound like you live here.