For my text to speech engine (I use Cereproc, William Voice). The engine can be used on the website here: https://www.cereproc.com/ and it supports the ssml tags used in this question..


I am building an acronymn interpreter that will clearly dilineate the acronymn as well as account for nuances like this:

IEDS -> I, E, Dees

instead of

IEDS -> I, E, Dee, Es

I only mention this, as this is why I can not use Cereproc's

<say-as interpret-as='characters'>IEDS</say-as>

Anyway, with phonemes, I am being tripped up here:

<phoneme alphabet="IPA" ph="ˈaɪ ˈiː dˈiːz ">IEDS</phoneme>

sounds something like "Iyee'Edees", giving a stupid sound artifact. I could do this:

 <phoneme alphabet="IPA" ph="ˈaɪ">I</phoneme>
 <phoneme alphabet="IPA" ph="ˈiː">E</phoneme>
 <phoneme alphabet="IPA" ph="dˈiːz">DS</phoneme>

But now the issue is the pause between each sound is too long.


Thus, in the IPA phonetic alphabet, what symbols could I place in between:


To give the pause I need to adequately space out those sounds so each letter is pronounced in full?

1 Answer 1


So in researching the question, I ended up finding an answer.

While it does not give me full control over the pausing, adding a period after each block of phonetic characters added the perfect amount of spacing, thus:

<phoneme alphabet="IPA" ph="  ˈaɪ. ˈiː. dˈiː.z ">IEDS</phoneme>

Works perfectly.

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.