How to Add Pauses in Google Text-to-Speech Using SSML

Example:

<speak>

This is the first part of text to speech

<break time="2s"/>

This is the second part of the text to speech

</speak>

<speak>  This is the first part. <break time="2s"/> This is the second part after a two-second pause. </speak>

Google Text-to-Speech Pauses Using SSML

Google Text-to-Speech (now part of Google Cloud Text-to-Speech) uses SSML (Speech Synthesis Markup Language) to control features like pauses.

There is no single “pause command” like a simple keyword. Instead, the standard way to insert a deliberate pause is with the <break> tag in SSML.

How to Add a Pause

Wrap your text in <speak> tags and insert <break time="Xs"/> (or milliseconds) where you want silence.

Examples

• Short pause:
  <break time="500ms"/> (half a second)

• Medium pause:
  <break time="1s"/> (1 second)

• Longer pause:
  <break time="3s"/> (3 seconds)

Full Example SSML Input

<speak>  This is the first part. <break time="2s"/> This is the second part after a two-second pause. </speak>

Strength-Based Pauses

You can also use strength-based pauses instead of specifying exact timing.

Options include:

• <break strength="weak"/>

• <break strength="medium"/>

• <break strength="strong"/>

• <break strength="x-strong"/>

These allow the speech engine to determine the exact pause length automatically.

Simple Alternatives (Without Full SSML)

If you are not using full SSML support, basic punctuation can still create natural pauses.

Examples include:

• Commas (,) – create short pauses

• Periods (.) – create natural sentence pauses

• Ellipses (...) – often produce a slightly longer breath-like pause

• Line breaks or new paragraphs – may create paragraph-level pauses depending on the implementation

Notes

This mainly applies to the Google Cloud Text-to-Speech API, which is used by developers, applications, and services such as Google Assistant integrations.

Some consumer tools (for example Android's built-in text-to-speech or Google Translate's read-aloud feature) may only support basic punctuation pauses and may not support full SSML markup.

The official documentation confirms that the <break> tag is the standard method for precise pause control in Google Text-to-Speech.