# Fish Audio Drama 3

Fish Audio Drama 3, a model in Sloosh, is text to speech you direct like an actor. You write [directions], sounds, pauses and per-word tags into the script, and it reads them as a performance. It uses the same 24 ready-made voices as Fish Audio S2.1 Pro and takes scripts up to 10,000 characters.

Web page: https://sloosh.ai/models/fish-drama-3

[Open Audio Studio](https://sloosh.ai/studio/audio)

## Key facts in Sloosh

|   |   |
| --- | --- |
| Made by | Fish Audio |
| Makes | speech audio |
| Inputs | script text, a voice, [directions], sounds, pauses, per-word tags |
| Voices | the same 24 ready-made voices as Fish Audio S2.1 Pro |
| Script limit | 10,000 characters |
| Cost | by script length: a 1,000-character script quoted 0.153 credits |
| Where | Audio Studio, Spacelab, MCP |

## What it is for

- Scenes and dialogue where each line needs its own delivery.
- Reads with sounds and pauses written into the script.
- Stressing single words with per-word tags.

## How to use Fish Audio Drama 3 in Sloosh

1. Open [Audio Studio](https://sloosh.ai/studio/audio), signed in.
2. On Voices, press a voice card to put it in the prompt box.
3. Pick Fish Audio Drama 3 in the model chip.
4. Write the script, set how many takes (up to 4), and check the cost on the Generate button before you press it.

## Directing a read

- Put a direction in square brackets before the line it shapes, such as [nervous] or [laughing].
- Write pauses and sounds into the script where they happen.
- Tag only the words that need it; a line full of tags is harder to read well.

**Script**

```text
[nervous] I was not going to say anything. [quietly] But you deserve to know. The ring was never lost.
```

## Drama 3 vs S2.1 Pro

Both are Fish Audio models with the same 24 voices and a 10,000-character limit. Sloosh's catalog lists Drama 3 for directed reads (directions, sounds, pauses, per-word tags) and [Fish Audio S2.1 Pro](https://sloosh.ai/models/fish-s2.1-pro) for text to speech in 83 languages with [bracket] directions and your own cloned voices. S2.1 Pro also has a Free version. Clone only a voice you have the right to use.

## Cost

The price depends on the length of the script. On 11 Nov 2026 a 1,000-character script quoted 0.153 credits. The Generate button shows the exact price before you run.

## Where you can use it

- Audio Studio: pick it in the model chip.
- Spacelab: add a Text to Speech node and choose Fish Audio Drama 3 in its Model picker.
- Claude or ChatGPT: connect [Sloosh MCP](https://sloosh.ai/docs/mcp.md) and ask for speech made with Fish Audio Drama 3.

Last checked 11 Nov 2026.

## Frequently asked questions

### What is Fish Audio Drama 3?

A Fish Audio text-to-speech model inside Sloosh that you direct like an actor, with [directions], sounds, pauses and per-word tags in the script.

### How is it different from Fish Audio S2.1 Pro?

Drama 3 is listed for directed reads; [S2.1 Pro](https://sloosh.ai/models/fish-s2.1-pro) is listed for 83 languages and your own cloned voices, and has a Free version. Both have the same 24 voices and 10,000-character limit. Clone only a voice you have the right to use.

### Which languages does it read, and how long can a script be?

Scripts can be up to 10,000 characters. Sloosh's catalog gives no language count for Drama 3, so for a set language list use [S2.1 Pro](https://sloosh.ai/models/fish-s2.1-pro).

### What does it cost?

It is priced by script length. On 11 Nov 2026 a 1,000-character script quoted 0.153 credits; the Generate button shows the exact price first.

### Where can I use it?

In Audio Studio, in Spacelab's Text to Speech node, and from Claude or ChatGPT through [Sloosh MCP](https://sloosh.ai/docs/mcp.md).

## Related

- Models: [Fish Audio S2.1 Pro](https://sloosh.ai/models/fish-s2.1-pro), [Gemini TTS](https://sloosh.ai/models/gemini-tts), [all models](https://sloosh.ai/models)
- More speech models: [Fish Audio S2.1 Pro](https://sloosh.ai/models/fish-s2.1-pro), [ElevenLabs v4](https://sloosh.ai/models/elevenlabs-v4), [MiniMax Speech 2.8 HD](https://sloosh.ai/models/minimax-speech-2.8-hd), [Gemini TTS](https://sloosh.ai/models/gemini-tts), [all models](https://sloosh.ai/models)
- Docs: [Audio Studio](https://sloosh.ai/docs/audio-studio.md), [Sloosh MCP](https://sloosh.ai/docs/mcp.md)

## Sources

- [Sloosh docs: Audio Studio](https://sloosh.ai/docs/audio-studio)
- [Sloosh docs: MCP](https://sloosh.ai/docs/mcp)