Features of connected speech

Words are ‘chunked’ together:

This sentences are divided into ‘chunks’ or strings of words with small pauses between them. For example, the sentence “It’s going to cost a lot of money, but it’s worth it.” would be divided into the following chunks: [It’s going to] [cost a lot of money] [but it’s worth it.]

Sounds change as they are influenced by sounds around them

Sounds change as they are influenced by sounds before them and sounds after them. For example, the /r/ in rink will be slightly different to the /r/ in brink, because it would take time to move the tongue from its position when we make the /b/ sound, to the position where it is put in the /r/ sound, and then to its position for the /ɪ/ sound. In the same way when words link together in connected speech, the last sound in a word will influence the first sound in the next word and that sound at the beginning of the next word will influence it. For example, in the sentence: “Do you know that boy over there?” the word that will not sound the way it would on its own or in another sentence. Normally, when you make the /t/ sound, your tongue briefly touches the roof of your mouth, just behind your top teeth. When speaking at normal speed, you need to quickly move from the /t/at the end of that to where it needs to be when you make the /b/ sound at the start of boy. The /t/ will be reduced and your tongue will not quite touch the roof of your mouth. However, if you are speaking in a formal situation (for example giving a public speech or oral presentation) you may be speaking more slowly and carefully in order to be easily understood. In that case, you are more likely to pronounce the two words separately and the /t/ will be fully produced.

Another common sound change is when a /t/ sometimes becomes a /d/ sound when followed by a vowel sound. Common examples are shut up, got it and water which may be pronounced as ‘shuddup‘, ‘goddit‘ and ‘warda“. In phonemic script these words look like /ˈʃʌd.ʌp/ /ˈgɒd.ət/ and /ˈwɔ:.də/.

Rhythm, intonation, stress and weak forms

English is not pronounced with flat intonation: it has a rhythm which rises and falls as some words are stressed and others are not stressed. Connected speech is largely a result of this rising and falling rhythm and stress. Stressed words are louder and longer, while unstressed words are softer and shorter. Many words have two forms; a ‘strong form’ and a ‘weak form’. When the word is unstressed the weak form will be used and when it is stressed, the strong form is used. Use of the strong or weak form depends on meaning.

Words have weak and strong forms:

As discussed in the previous paragraph, many words in English have two pronunciation forms – a ‘weak form’ and a ‘strong form’. You are probably familiar with the weak/reduced and in the phrase fish and chips, which is sometimes written as fish n chips to emphasise the weak form of and.  Some very common examples are:

	Strong	Weak
was	/wɒz/	/wəz/
were	/wɜ:/	/wə/
just	/ʤʌst/	/ʤəst/ or /ʤəs/

The weak form is used when the words are unstressed in a sentence and the strong form is used when they are stressed.

“Using the strong or weak form depends on meaning. Choosing to use the strong form will change the meaning of the sentence”.

Click here for more information about rhythm, stress and weak/strong forms.

Click here for more examples of words with strong/weak forms

When we use weak and strong forms

Use of special stress to emphasise, contradict and communicate precise meaning

The meaning of a sentence can be dramatically changed by placing special stress on one word. When a word is stressed, it cannot be weak and must be produced in its strong form. When placing special stress on a word, it is usual that the stressed word won’t connect to the words before and after and will be pronounced as a separate unit.

Words join together or ‘contract’:

You will be very familiar with more common contractions such as can’t, won’t, don’t and haven’t. These contractions have existed for quite a long time and are acceptable in many social situations in speaking and are also accepted in many written text-types. However, there are also many newer and less formal contractions which are not as accepted and which should be only used in very informal settings. In the written form you are only likely to see them in SMS messages, online chats and comic books. Some examples are don’t know > dunno, got you > gotcha and want to > wanna.

Click here for more information about contractions

Sounds are dropped:

Sometimes we drop sounds because two sounds require similar use of vocal apparatus and it is quicker to drop a sound. For example, next time might be reduced to ‘nex time‘ because pronouncing the two /t/ sounds would require a slight pause to separate the two words and more effort. Similarly, because /t/ and /d/ are similar sounds, a /t/ is often reduced or dropped when the next word starts with a /d/.

Examples

	Sentence	May sound like
1	See you next time.	See you nex time.
2	I must do that.	I mus do that.
3	I just phoned him five minutes ago.	I jus phoned ‘im five minutes ago.
4	They walked to school.	They walk to school.
5	brand new	bran new

In Example 3., the /t/ in just and the /h/ in him have been dropped. When a word starts with a /h/, it is often dropped in informal English and in some dialects in some dialects in England. Generally dropping the /h/ is an informal or dialectal phenomenon, but it can affect connected speech in a sentence. In this example, dropping the /h/ helps maintain the flow of speech as it allows phoned to connect to ‘im. ‘im here is a weak form of him.

The two sentences in Example 4., will sound identical when spoken at the normal speed because repeating the two /t/ sounds (walked & to) would disrupt the smooth flow of speech.

Sounds are inserted:

Linking and added sounds. Sometimes a sound is inserted in order to link two words smoothly. A combination of a word ending in a vowel followed by a word starting with a vowel can cause that flow to be interrupted. For example, in the phrase I saw a dog, the /ɔ:/ sound followed immediately by the /ə/ sound in a is difficult to say, so in some dialects of English, we automatically insert an /r/ sound to bridge across the two words. Click here for more information and listening exercises.

Unstressed component words in compound words are reduced

When we create compound nouns, we usually stress the first word in the compound noun. With a small group of high-frequency compound words containing man/men & land we also reduce the second word. This reduction of the second word is an element of connected speech. Consider these examples:

As you can see in the table above, the singular and plural forms of postman and postmen are pronounced the same. This happens with other compounds containing the word man/men, such as workman/workmen and tradesman/tradesmen. Generally, in compound words ending with land, the ending is pronounced /lənd/as in the examples above. Some other examples are country names such as Scotland, Ireland and others and placenames such as Sunderland. However, there are exceptions in which the ending is pronounced /lænd/. Some examples are hinterland, wonderland, and badlands.

With some words, the way syllables break up does not correspond to the original words. Consider bedroom, which may be pronounced /ˈbed.ruːm/ or /ˈbe.druːm/