---
title: Inline tags — Gemini 3.8 Flash TTS | Runware Docs
url: https://runware.ai/docs/models/google-gemini-3-8-flash-tts/guides/inline-tags
description: "How to add laughs, sighs, breaths and pauses to Gemini 3.8 Flash TTS speech with inline tags: the syntax, the documented tags and where to place them."
---
### [Introduction](https://runware.ai/docs/models/google-gemini-3-8-flash-tts/guides/inline-tags#introduction)

Inline tags add **a sound or a pause at an exact point** in the speech. You write them in angle brackets inside `speech.text`, such as `<laugh>` or `<short pause>`, and the model **performs them instead of reading them**: it makes the sound or leaves the silence where the tag sits, then carries on with the text.

[Listen to audio](https://runware.ai/docs/assets/hero.CMLhsFI7.mp3)

*Callirrhoe*

> **Prompt**: <yawn> Morning, everyone. Okay... coffee first. <chuckle> Then I'm finally showing you the studio we spent all summer building.

This guide covers the request, the documented tags, pauses, breath and effort, laughs and reactions, tags with a style and tags in other languages.

> [!NOTE]
> Gemini 3.1 Flash TTS writes these tags in square brackets. Gemini 3.8 Flash TTS uses angle brackets, and square brackets mark the speakers of a [dialogue](https://runware.ai/docs/models/google-gemini-3-8-flash-tts/guides/dialogue).

### [The request](https://runware.ai/docs/models/google-gemini-3-8-flash-tts/guides/inline-tags#the-request)

A tag goes **exactly where the sound happens**, between the words around it, and the text after it continues in the same delivery.

**TypeScript**:

```typescript
import { createClient } from '@runware/sdk'

const client = await createClient({ apiKey: process.env.RUNWARE_API_KEY })
await client.connect()

const [result] = await client.run({
  model: 'google:gemini@3.8-flash-tts',
  speech: {
    text: '<yawn> Morning, everyone. Okay... coffee first. <chuckle> Then I\'m finally showing you the studio we spent all summer building.',
    voice: 'Callirrhoe'
  }
})
```

**Python**:

```python
import asyncio
import os

from runware import Runware

async def main():
    async with Runware(api_key=os.environ["RUNWARE_API_KEY"]) as client:
        results = await client.run({
            "model": "google:gemini@3.8-flash-tts",
            "speech": {
                "text": "<yawn> Morning, everyone. Okay... coffee first. <chuckle> Then I'm finally showing you the studio we spent all summer building.",
                "voice": "Callirrhoe"
            }
        })

asyncio.run(main())
```

**cURL**:

```bash
curl https://api.runware.ai/v1 \
  -H "Authorization: Bearer $RUNWARE_API_KEY" \
  -H "Content-Type: application/json" \
  -d '[
    {
      "taskType": "audioInference",
      "taskUUID": "2b1084c3-4780-4106-882a-948456fda422",
      "model": "google:gemini@3.8-flash-tts",
      "speech": {
        "text": "<yawn> Morning, everyone. Okay... coffee first. <chuckle> Then I'm finally showing you the studio we spent all summer building.",
        "voice": "Callirrhoe"
      }
    }
  ]'
```

**CLI**:

```bash
runware run google:gemini@3.8-flash-tts \
  speech.text="<yawn> Morning, everyone. Okay... coffee first. <chuckle> Then I'm finally showing you the studio we spent all summer building." \
  speech.voice=Callirrhoe
```

**JSON**:

```json
{
  "taskType": "audioInference",
  "taskUUID": "2b1084c3-4780-4106-882a-948456fda422",
  "model": "google:gemini@3.8-flash-tts",
  "speech": {
    "text": "<yawn> Morning, everyone. Okay... coffee first. <chuckle> Then I'm finally showing you the studio we spent all summer building.",
    "voice": "Callirrhoe"
  }
}
```

### [Documented tags](https://runware.ai/docs/models/google-gemini-3-8-flash-tts/guides/inline-tags#documented-tags)

Google lists the tags below for the model, grouped here by kind. They cover **sounds a person makes**, so sound effects such as applause or a door knock are outside the set.

```text
Pauses     <short pause> <long pause>
Breath     <breath> <heavy breath> <exhales> <pant> <phew> <pff>
Laughter   <laugh> <laughter> <chuckle> <chuckles> <giggle> <snicker> <cackle>
Reactions  <sigh> <sighs> <gasp> <groan> <tsk> <snort> <argh> <grr> <growl> <grunt> <hiss>
Voice      <whispers> <whispering> <shout> <scream> <shriek> <cheer>
Body       <cough> <sneeze> <yawn> <throat-clearing>
Distress   <cry> <sob> <whimper> <moan>
```

### [Pauses](https://runware.ai/docs/models/google-gemini-3-8-flash-tts/guides/inline-tags#pauses)

`<short pause>` and `<long pause>` leave **a silence where the listener expects a beat**, such as before a reveal or after a question. Punctuation paces the read as well, as [Prompting](https://runware.ai/docs/models/google-gemini-3-8-flash-tts/guides/prompting) covers, and a tag makes the pause part of the performance.

[Listen to audio](https://runware.ai/docs/assets/pause-long.BSnZag2n.mp3)

*Fenrir*

> **Prompt**: And this week's top scorer, with four thousand two hundred points, is... <long pause> Priya!

A voice agent can use **a short pause between two pieces of information**, so the customer hears the total before the question.

[Listen to audio](https://runware.ai/docs/assets/pause-short.DfFPhUlz.mp3)

*Achird*

> **Prompt**: Your total comes to forty-two fifty. <short pause> Would you like the receipt sent to your email?

### [Breath and effort](https://runware.ai/docs/models/google-gemini-3-8-flash-tts/guides/inline-tags#breath-and-effort)

Breath tags make **physical effort audible**, which suits workout coaching and sports content, where a perfectly smooth read sounds like it was recorded sitting down.

[Listen to audio](https://runware.ai/docs/assets/breath.YdL6WQio.mp3)

*Alnilam*

> **Prompt**: Last round. <heavy breath> Ten more squats, chest up, push through your heels. <exhales> Good. Five, four, three, two... and done.

### [Laughs and reactions](https://runware.ai/docs/models/google-gemini-3-8-flash-tts/guides/inline-tags#laughs-and-reactions)

A reaction tag puts **a response into the speaker's voice**, such as surprise at an unboxing or a groan at bad news.

[Listen to audio](https://runware.ai/docs/assets/reaction-gasp.CVWXChhc.mp3)

*Zephyr*

> **Prompt**: <gasp> Okay, it's even smaller than I thought. <laugh> It actually fits in my jacket pocket.

[Listen to audio](https://runware.ai/docs/assets/reaction-sigh.CX3ML8CK.mp3)

*Umbriel*

> **Prompt**: <sigh> So I finally checked my screen time. Six hours. <groan> On a Tuesday.

Place a reaction **where a person would make it**. A sigh opens a resigned line, and a laugh follows the line it reacts to.

### [Combining tags with a style](https://runware.ai/docs/models/google-gemini-3-8-flash-tts/guides/inline-tags#combining-tags-with-a-style)

A tag and a style do different jobs in the same request. `settings.style` sets **the mood of the whole read**, and the tags mark the moments inside it.

[Listen to audio](https://runware.ai/docs/assets/with-style.CaQDAkqw.mp3)

*Despina. Style: relieved and a little giddy, late at night*

> **Prompt**: <exhales> Okay. It's live. <chuckle> We actually shipped it. Thank you, all of you, for staying up with us tonight.

[Delivery style](https://runware.ai/docs/models/google-gemini-3-8-flash-tts/guides/delivery-style) covers writing the direction.

### [Tags in other languages](https://runware.ai/docs/models/google-gemini-3-8-flash-tts/guides/inline-tags#tags-in-other-languages)

**Keep the tags in English** whatever the language of the text. The words around them change, and the tag stays as it is in the list above.

[Listen to audio](https://runware.ai/docs/assets/other-language.DxZhjxRr.mp3)

*Puck, Spanish*

> **Prompt**: <laugh> Vale, lo reconozco: la primera vez que monté la estantería me sobraron seis tornillos. <sigh> Y todavía no sé de dónde salieron.

### [Tips](https://runware.ai/docs/models/google-gemini-3-8-flash-tts/guides/inline-tags#tips)

1. **Put the tag where a person would make the sound.** Before the line for a sigh, after the joke for a laugh.
    
2. **Use the documented tags.** They are the ones Google describes for the model, and they all stand for a sound a person makes.
    
3. **Keep tags in English.** The text can be in any language the model reads.
    
4. **Use a style for a mood.** A tag is one moment, and a delivery that lasts belongs in `settings.style`.