How to Pronounce -ED Endings: /t/, /d/ and /ɪd/ Rules

The -ed ending has three pronunciations: /t/, /d/ and /ɪd/. Which one you use depends on the final sound of the base verb, not on its spelling. Walked is /wɔːkt/, played is /pleɪd/, and wanted is /ˈwɒntɪd/.

Only one of the three adds a syllable. Walked and played stay at one syllable each, and adding a second is the single most common error with this ending. Type any past-tense verb into the pronunciation Studio to hear which of the three it takes.

The rule in one table

Check the last sound of the base verb, then read across.

That is the whole system. Check /t/ and /d/ first, because they override everything else, then decide voicing for the rest.

The voiceless sounds that take /t/ are /p/, /k/, /f/, /s/, /ʃ/, /tʃ/ and /θ/. Everything else in English is voiced: every vowel, and the consonants /b/, /ɡ/, /v/, /ð/, /z/, /ʒ/, /dʒ/, /m/, /n/, /ŋ/, /l/, /r/ and /w/.

If voicing is unfamiliar, put two fingers on your throat and hold a long “sss”, then “zzz”. The buzz that appears on the second is voicing, and the voiced and voiceless consonants guide covers the full set of pairs.

Sound, not spelling

This is where most errors come from. The rule looks at the last sound, and English spelling regularly disagrees with it.

A verb ending in a silent e is the clearest case. Like ends in the letter e and the sound /k/, so it takes /t/. Reading the spelling would give you /ɪd/, which is wrong.

The other trap is a verb that ends in the letter t or d but not the sound. There are very few, but doubt is one worth knowing: the b is silent, the word ends in /t/, and doubted is /ˈdaʊtɪd/, which the rule handles correctly.

Why the rule exists at all

It helps to know that this is not a convention somebody decided on. It is a consequence of how the mouth works, and seeing that makes the rule much easier to trust.

Voicing is a switch at the vocal folds. Turning it on and off takes a measurable amount of time, so a language that demanded a voiced consonant immediately after a voiceless one would be asking speakers to flip the switch in the middle of a closure. English does not ask that. It lets the ending inherit the setting that is already there.

Say walked and pay attention to the /k/. Your voice is already off for it. The /t/ that follows simply continues with the voice off. Now say played. The vowel before the ending is voiced, your voice is already on, and the /d/ continues with it on. In both cases the ending is whichever version costs nothing.

The /ɪd/ case is a different problem with the same kind of solution. After a /t/, an added /t/ would be inaudible, because two identical stops in a row sound like one slightly longer stop. English needs the past tense to be hearable, so it inserts a vowel to separate them. The same thing happens after /d/.

That is why the three groups are not equally sized, and why the odd one out is the one that gains a syllable. Two of the groups are solving a voicing problem and the third is solving an audibility problem.

Group 1: -ed said as /ɪd/

Only after /t/ or /d/. This is the one group where -ed becomes a separate syllable, and it exists because English cannot put two t sounds or two d sounds directly together and expect anyone to hear them.

Some speakers and some dictionaries write this ending /əd/ rather than /ɪd/, since the vowel is weak and varies. Both notations describe the same thing, and this site uses /ɪd/ consistently.

Group 2: -ed said as /t/

After any voiceless sound except /t/.

Every one of those is a single syllable, apart from finished, which has two because the base verb already did.

Some of them end in a demanding cluster. Asked is /ɑːskt/, with three consonants after the vowel. That is a cluster problem rather than an -ed problem, and the consonant clusters guide covers how English handles those runs.

Group 3: -ed said as /d/

After any voiced sound except /d/, including every vowel.

Changed is the interesting one. The base ends in /dʒ/, which contains a d sound, but /dʒ/ is not /d/, so the /ɪd/ rule does not apply. It is voiced, so the ending is /d/, and the word stays at one syllable: /tʃeɪndʒd/.

The adjectives that break the rule

About a dozen words ending in -ed keep an extra syllable when used as adjectives, even though the rule says they should not. These are leftovers from an older stage of English.

A learned professor has three syllables in learned. She learned French has one. The difference is grammatical, not phonetic.

What happens to -ed in a real sentence

The rule above describes the word said on its own. In running speech, the /t/ and /d/ endings often disappear, and this is normal rather than sloppy.

A final /t/ or /d/ at the end of a consonant cluster drops when the next word starts with a consonant. That applies to the past tense like anything else.

The ending comes back when the next word begins with a vowel, because there is now somewhere for it to go: walked away keeps its /t/ and links it across, giving /wɔːk təˈweɪ/.

Two things follow from this. First, if you cannot hear a past tense in fast speech, that may be because it genuinely is not there as a separate sound, and the tense is carried by context instead. Second, dropping it yourself is fine in conversation but risky while you are learning, because dropping the wrong sound produces a word that is simply unclear. Get the full form solid first, and let the reduction happen on its own. The connected speech guide covers the wider set of processes at work here.

The /ɪd/ ending never drops. It is a whole syllable, and syllables do not vanish the way single consonants do.

The three mistakes people actually make

Adding a syllable everywhere. Walked said as WORK-ed, played said as PLAY-ed. This comes from reading the spelling, and it is by far the most common error. Both words are one syllable.

Dropping the syllable where it belongs. Wanted said as WANT. Here the syllable is compulsory, because without it the past tense is invisible.

Using /d/ after a voiceless sound. Laughed said as /lɑːfd/. Try it and you will notice your mouth resisting: producing a voiced /d/ straight after a voiceless /f/ takes deliberate effort, which is why English does not do it.

The third mistake is worth dwelling on, because it shows the rule is not arbitrary. English is not choosing /t/ after voiceless sounds as a convention. It is choosing the one that the mouth can produce without stopping.

Practice

Say each verb in its past form, then decide which of the three endings it takes. Cover the answers first.

1. stopped
2. needed
3. cleaned
4. watched
5. decided
6. agreed
7. laughed
8. travelled
9. counted
10. finished
11. changed
12. asked
13. enjoyed
14. hated
15. pushed

If you got 11 and 4 right, the rule has landed. Those are the two where the spelling most strongly suggests the wrong answer.

Second practice: full sentences

Reading single words is easier than reading sentences, because a sentence gives your old habits somewhere to hide. Read each of these aloud at normal speed.

1. She watched the film and enjoyed it.
2. They wanted more time but we finished early.
3. He asked me twice and I agreed.
4. The train stopped and everybody waited.
5. I cleaned the kitchen, painted the door and hated every minute.

Sentence 5 is the useful test. Two /ɪd/ endings sitting close to a /d/ ending is exactly the situation where people start applying one ending to everything.

A quick way to check any verb

If you are unsure about a verb and have no reference to hand, this takes about five seconds.

1. Say the base verb out loud and stop at the last sound. Ignore the spelling completely.
2. Is that sound a /t/ or a /d/? If yes, the ending is /ɪd/ and the word gains a syllable. Stop here.
3. If not, put your fingers on your throat and make that last sound on its own, held.
4. Buzzing means voiced, so the ending is /d/. Silence means voiceless, so the ending is /t/.
5. Either way, the word does not gain a syllable.

The one step people skip is the first. Saying the verb rather than looking at it is what makes the difference, because the whole rule is about sound and the spelling is the thing leading you astray.

The same logic elsewhere

This is not a rule invented for the past tense. It is a general pattern in English: an ending takes the voicing of the sound before it.

The plural and third-person -s works exactly the same way, with /s/ after voiceless sounds, /z/ after voiced ones, and /ɪz/ after the sounds too similar to hear next to it. That system is covered in the plural -s guide, and the two rules are worth learning together because they share one idea.

To check a verb you are unsure about, type it into the Studio and read the transcription. If the ending appears as /ɪd/, the word gained a syllable. If it appears as a bare /t/ or /d/, it did not.