Intonella

Connected speech: why English doesn’t sound the way it’s written.

In a sentence, English words don’t stand apart. They join, shrink and lose sounds. That is connected speech, and it is why you can read a sentence easily and still miss it when someone says it.

Ben, in “At the farmers' market”

Written
Then let's geta dozen. And whatabout cheese?
Heard
then lets geda dozen. n whadabout cheese?

Four changes that do most of it

Each example below is a line from the same dialogue. Play it and listen for the change.

Linking

When a word ends in a consonant and the next one starts with a vowel, the consonant moves over. “Make it” sounds like “may kit”. You hear one word, not two.

  • Written
    Isn'tit?
    Heard
    izznit?
  • Written
    littleofeach
    Heard
    liddle uh veach

The flapped t

Between two vowel sounds, American English turns t into a quick, soft d. “Saturday” becomes “Saddurday”, and “get a” becomes “geda”.

  • Written
    some for Saturday?
    Heard
    some fer Saddurday?
  • Written
    geta dozen.
    Heard
    geda dozen.
  • Written
    whatabout cheese?
    Heard
    whadabout cheese?

Weak forms

Small words like of, for, and and to lose their vowel and almost disappear. You hear the big words clearly; the small ones are a murmur between them.

  • Written
    halfa poundof that,
    Heard
    haffa pounda that,
  • Written
    and a loafof sourdough.
    Heard
    n a loafa sourdough.

Dropped sounds

When consonants pile up between words, one of them often goes. “Taste these” sounds like “tase these”. The t is in the spelling, not in the sound.

  • Written
    taste these two?
    Heard
    tase these two?
  • Written
    bringsher sister.
    Heard
    bringzer sister.

The heard lines are spelled the way learners’ books write connected speech. They show what American speakers usually do. Not every speaker does every one, every time, so listen for them.

You learn it to hear, not to sound like a textbook.

Once you know that “what about” arrives as “whadabout”, you stop waiting for two words that never come. But knowing the rules is the small part. The big part is hearing the same joins many times, in real sentences, until your ear does it without you.

How to train your ear for it

The same three steps, on any short recording. Use the lines on this page to start.

How shadowing works in Intonella
  1. Hear it before you read it.

    Play a short line many times with no text. Your ear starts to find the words inside the joins.

  2. Then read along.

    With the text in front of you, you see where the words joined and which small ones went quiet.

  3. Then say it with the speaker.

    Saying the joins yourself is the fastest way to hear them. Speak with the recording, at the same time and at its speed.

Every dialogue in the app is full of it.

Intonella’s dialogues are everyday American English at native speed, about a minute each. You hear one with no text, then read along as each line lights up, then say it with the speakers. The joins on this page are in every one of them.

Read along: the dialogue’s text, with the line being spoken lit up.Speak along: the text with the line to say lit up, played a little slower.

Questions about connected speech

What is connected speech?

How words change when they are spoken together instead of one by one: they link, shrink and lose sounds. Every native speaker does it, careful speakers too.

What are the main features of connected speech?

Linking, weak forms, dropped sounds (elision), sounds that change to fit the next one (assimilation), and sounds added between words (intrusion). In American English the flapped t is one of the most common.

Is connected speech lazy or wrong?

No. It is how fluent English sounds. Newsreaders and teachers do it too. Saying every word separately sounds slow and stiff to a native ear.

Do I need to use connected speech when I speak?

You will be understood without it. But saying the joins yourself is the quickest way to start hearing them, and your English will sound more relaxed.

How long does it take to hear it?

Weeks, not years. Each dialogue you learn well trains your ear for the same joins in every other sentence.

One dialogue. A few days. Hear the difference.

Get the app

For iPhone and Android.