How to Adjust Pauses and Silence in Ondoku Speech【2 Methods】
Nov. 12, 2026
Hello, and thank you for always using Ondoku.
One need among Ondoku users is “I’d like a little more space between phrases.”
We don’t usually pay much attention to pauses in everyday conversation, but when you have Ondoku read text aloud, you realize just how important they are.
Whoever masters the pause masters the world.
This time, we’ll show you how to add pauses.
If you want to adjust the “pause” to leave a little more space, there are two ways to do so:
- Punctuation
- SSML
※Pause adjustment is supported only in the original version.
Punctuation and SSML (breaktime tags) cannot be used in the Beta version. Please note that they may cause errors.
Adjusting pauses with punctuation
You can adjust pauses slightly using punctuation such as
- ,
- .
- …
.
However, this does not let you create a “precise pause of ◯ seconds.”
Punctuation can only adjust pauses by about the length of a single breath.
To adjust pauses,
- “,,,,,,,,”
- “........”
- “……………………”
some people try entering punctuation repeatedly, but using multiple punctuation marks has no effect.
Ondoku is not designed to adjust pauses based on the number of punctuation marks.
Repeated small “っ” characters, as in “っっっっっ,” can have a slight effect on lengthening the pauses.
However, methods that use lots of punctuation or small “っ” characters cannot create precise pauses, and they use up a large number of characters from your quota.
When you want a longer pause or a precise interval of silence, use SSML tags.
How to create a precise pause of a specified number of seconds (SSML tags)

To create a precise pause, we recommend
using SSML tags to adjust the pause.
If you use this method, keep the number of characters read aloud at one time under 3,000. Text of 3,000 characters or more is more likely to cause an error.
What is SSML?
SSML is a markup language for speech. HTML is a well-known example of a markup language.You might wonder, “Markup something you can’t see?” but it lets you control how machine-generated speech is spoken.
SSML is written similarly to HTML, so if you’ve ever created a website, you may find it easy to understand.
Learn more about SSML in this article.
To add pauses with SSML, write the tags as shown below.
To make the pauses easy to understand, we’ve included pauses of 1–5 seconds.
<speak>
S stands for seconds<break time="1s"/>in English<break time="2s"/>
Increasing the number before S makes the pause at that point longer<break time="5s"/>
MS<break time="500ms"/>stands for milliseconds
</speak>
Only two tags are needed to create pauses.
- <speak></speak>
- <break time="2s"/>
<speak></speak>
This tag is required to tell Ondoku, “This text uses SSML.”
Add it at both the beginning and end of the text.
<break time="○s"/>
This tag means “break time” or “pause.”
Enter your preferred number in place of ○ to create a pause of that duration.
s is short for seconds.
1s means 1 second.
ms stands for milliseconds.
“1000ms = 1s,” so 1,000 milliseconds is the same as 1 second.
※The maximum pause you can specify with an SSML tag is 10 seconds. If you set a longer duration, it will be rounded down to 10 seconds.
The SSML tags currently supported in all languages are
- <speak></speak>
- <break time="○○ms"/>
these two types only. Other tags may not be supported, depending on the language and voice type. Thank you for your understanding.
※The behavior of a <break time="○○ms"/> tag placed before the text varies by voice and is not guaranteed.

<speak></speak>
<break time="○○ms"/>
These two tags do not count toward the character count.
However, if there is even a single error, such as an extra space inside a tag, it will not be recognized correctly. In that case, the tag will count toward the character count.

Notes on using SSML
- Keep the text under 3,000 characters
- Do not use unnecessary symbols
- Even a one-character error in a tag will prevent the text from being read aloud
- Do not specify an excessively long duration
- Do not use a <break time="○○ms"/> tag before the text
- Supported only in the original version. Not supported in the Beta version
1. Keep the text under 3,000 characters
If you use SSML, keep the number of characters read aloud at one time under 3,000. Text of 3,000 characters or more will cause an error.
2. Do not use unnecessary symbols
When using SSML, make sure the text to be read aloud does not contain symbols.
In particular, < > 「」& are symbols that cannot be used, so if they appear in the text you want read aloud, the text will not be read.
Please also remove any other symbols before reading the text aloud.
3. Even a one-character error in a tag will prevent the text from being read aloud
Like programming code, SSML will not work if there is even a one-character error.
This includes extra spaces or slashes in a tag, or even a single mistake in the spelling of the letters.
Even a small error like that will prevent the text from being read aloud.
We recommend copying and pasting tags from the website.
Tags are very delicate. Be careful when writing them yourself.
4. Pauses and silences can be set for up to 10 seconds
With Ondoku, the maximum pause or silence you can specify using an SSML tag is 10 seconds.
If you specify a longer duration, it will be read as a 10-second pause or silence.
5. Do not use a <break time="○○ms"/> tag before the text
The behavior of a <break time="○○ms"/> tag placed before the text varies by voice and is not guaranteed.
Example: <speak><break time="5s"/>I want to add a five-second pause at the beginning</speak>
↑This use is not supported.
6. Supported only in the original version. Not supported in the Beta version
The SSML breaktime tag is supported only in the original version of Ondoku.
The breaktime tag cannot be used in the Beta version.
For other errors that may occur when using SSML, see here.
We’ve introduced two ways to add pauses to speech read aloud by Ondoku.
SSML may look difficult at first, but it’s very easy to use. If all you want to do is add pauses, it’s the easiest method and our top recommendation.
Give it a try!
■ AI voice synthesis software "Ondoku"
"Ondoku" is an online text-to-speech tool that can be used with no initial costs.
- Supports 81 languages and regions, including Japanese, English, Chinese, Korean, Spanish, French, and German
- Available from both PC and smartphone
- Suitable for business, education, entertainment, etc.
- No installation required, can be used immediately from your browser
- Supports reading from images
To use it, simply enter text or upload a file on the site. A natural-sounding audio file will be generated within seconds. You can use voice synthesis up to 5,000 characters for free, so please give it a try.
Email: ondoku3.com@gmail.com
"Ondoku" is a Text-to-Speech service that anyone can use for free without installation. If you register for free, you can get up to 5000 characters for free each month. Register now for free

