Add pauses and style with tags
Use tags in your script to add pauses, emotion and sounds. Here's what each model can do.
Updated November 10, 2026 · 1 min read
On this page
Add a pause
Type [break 1.5] where you want a pause of 1.5 seconds. Pauses work on every model and can be from 0.1 to 5 seconds long.
Add style and sounds
Type [ in the editor to see the tags your model supports. Use the arrow keys and Enter to insert one, or pick one from the Tags row above the script.
| Model | What it can do |
|---|---|
| Vega V2 | Any tag you write is read as a style. Some that work well: [happy], [whispering], [dramatic], [narrator], [laughing]. Use [normal] to go back to the usual delivery. |
| V3 Turbo | Sounds: [laugh], [chuckle], [sigh], [gasp], [cough], [clear throat], [sniff], [groan], [shush] |
| Vega V1, V1 (Classic), V2 (Multi-lingual) | Pauses |
| Fast Speech | Pauses |
For example, in Vega V2:
[warm] Welcome back to the show.
[break 1]
[whispering] I've got a secret to share.
If a tag is blocked
Tags your model can't perform are highlighted, and generating is blocked until you fix them. Click Remove tag to delete them, or switch to the suggested model.
Tags and credits
Short tags usually don't count towards your credits, and neither do pauses. The cost on the button is what you'll pay, so check it there.
Text that's cleaned up
Before generating, SpeakSay removes things a voice can't read, such as emoji, list markers and some symbols. If a sentence sounds odd, check your punctuation.
Getting a word to sound right
There isn't a special code for pronunciation. If a name or word sounds wrong, try spelling it the way it sounds, or try another model.