> ## Documentation Index
> Fetch the complete documentation index at: https://docs.comfy.org/llms.txt
> Use this file to discover all available pages before exploring further.
# Fish Audio Drama 3: directions, word tags, and templates
> Run the Fish Audio Drama 3 model in ComfyUI: direct a performance with plain-language directions and paired word tags, and use the ready-made Drama 3 workflow templates.
Drama 3 is Fish Audio's newest text-to-speech model, built for voice direction. A whole scene is written as one script, the delivery is described in plain language inside the text, and paired word tags change only the words they wrap. In ComfyUI it runs in the Fish Audio Text to Speech node with `model` set to `drama-3`. For the other Fish Audio models, see [Fish Audio: Text to Speech, Voice Cloning, and Speech to Text](/tutorials/partner-nodes/fishaudio/fish-audio).
To use the Partner Nodes, you need to ensure that you are logged in properly and using a permitted network environment. Please refer to the [Partner Nodes Overview](/tutorials/partner-nodes/overview) section of the documentation to understand the specific requirements for using the Partner Nodes.
Make sure your ComfyUI is updated.
* [Download ComfyUI](https://www.comfy.org/download)
* [Update Guide](/installation/update_comfyui)
Workflows in this guide can be found in the [Workflow Templates](/interface/features/template).
If you can't find them in the template, your ComfyUI may be outdated.
If nodes are missing when loading a workflow, possible reasons:
1. You are not using the latest ComfyUI version (Nightly version)
2. Some nodes failed to import at startup
* [Cloud](https://cloud.comfy.org) will update after ComfyUI stable release.
So, if you find any core node missing in this document, it might be because the new core nodes have not yet been released in the latest stable version. Please wait for the next stable release.
## Use Drama 3 in ComfyUI
The Fish Audio partner nodes are available in the node menu under `partner/audio/Fish Audio`, and Drama 3 uses the same ones as the other models: Text to Speech, Speech to Text, Voice Selector, and Instant Voice Clone.
### Selecting the model
Set `model` to `drama-3` on the Text to Speech node.
### Connecting voices
The `voice` output of the Voice Selector or the Instant Voice Clone connects to the Text to Speech node. `drama-3` exposes the growable `voices` input with up to 5 slots. With one connected voice, the node passes it internally as `reference_id`. With two or more, the text must mark speakers with `@Voice1`, `@Voice2`, and so on, and every connected voice must appear in the text at least once or the node raises an error.
## Directing a performance
`drama-3` takes direction from the script itself. A plain-language direction in square brackets shifts the delivery from that point on, and a paired word tag changes only the words it wraps:
```text theme={null}
[calm and inviting] Welcome back to the studio.
[leans in, conspiratorial] There is one trick nobody tells you about.
[light, playful] Ready? It is all in the timing. [bright, back to full voice] And that is the whole secret.
```
### Describe the delivery
Put a direction in square brackets before the words it applies to, and write it the way you would brief a voice actor: emotion, intent, pacing, or who the character is talking to.
```text theme={null}
[barely holding it together, forcing a smile] I'm fine. Really. Go on without me.
[professional broadcast tone] And now, the headlines.
```
A direction applies until the next one. `[calm]` at the start of a line holds for the whole line unless another direction replaces it, and several directions in one line shift the delivery within that line.
### Add sounds and pauses
Write a sound or a pause in brackets at the point where it should happen, such as a sigh, a laugh, or a breath:
```text theme={null}
I'm fine. [shaky breath] Really. [small laugh] Go on without me.
```
### Direct in any language
Directions do not have to match the language of the script. An English script can be directed in Chinese, and the other way around:
```text theme={null}
[像是在强忍着眼泪，却努力笑着] I'm fine. Really. Go on without me.
```
### Change only specific words
Wrap words in a paired tag to change the delivery of those words only. The supported tags are ``, ``, ``, ``, ``, ``, and ``:
```text theme={null}
It is going to blow.
I'm fine. Really. Go on without me.
```
### Switch language mid-line
Fish Audio lists language tags as experimental.
Wrap a phrase in a language tag to switch language for those words only. This helps in a line that mixes two languages, or when a phrase should be read in another language:
```text theme={null}
She looked at the menu and said Je voudrais une table pour deux, s'il vous plaît. The waiter smiled.
```
### Combine them
Directions, word tags, and the [phoneme tags](/tutorials/partner-nodes/fishaudio/prompt-guide#phoneme-control) all work in the same script, including inside a multi-speaker turn:
```text theme={null}
@Voice1 [calm and inviting] Welcome back to the studio.
@Voice2 [leans in, conspiratorial] There is one trick nobody tells you about.
@Voice1 [light, playful] Ready? It is all in the timing. [bright, back to full voice] And that is the whole secret.
```
The emotion, tone, and effect tags in the [prompt guide](/tutorials/partner-nodes/fishaudio/prompt-guide) belong to the other models. On `drama-3`, use the directions and word tags on this page instead.
## Available workflows
Two ready-made templates cover Drama 3: a single directed performance and a two-speaker dialogue. Open them from the Template Library in ComfyUI, or run them directly on Comfy Cloud. Both require ComfyUI 0.39.3 or newer.
Fish Audio Drama 3: Directed Performance
Act out one script with a single voice. Directions in the `text` field shift the delivery line by line, and `...` changes a few words without changing the rest of the line. The template starts from the Delia preset voice, and you can swap it in the Voice Selector.
Open in Comfy Cloud
Download JSON or search "Fish Audio Drama 3: Directed Performance" in Template Library
**Input materials**
Upload this file to the matching `LoadAudio` node:
`LoadAudio` node 11 · `fish_audio_example.mp3`
Fish Audio Drama 3: Two-Speaker Dialogue
Generate a whole scene in one pass, with one Voice Selector per speaker and the dialogue tags in the script. The template ships with the Delia and Sheila preset voices:
```text theme={null}
[deep, cinematic narrator] @Voice1 Every night, the city tells a story.
[soft, close, like a secret] @Voice2 But only one voice ever finishes it.
[drops to a whisper, then brightens] @Voice1 Fish Audio Drama 3, now in ComfyUI.
```
Open in Comfy Cloud
Download JSON or search "Fish Audio Drama 3: Two-Speaker Dialogue" in Template Library
**Input materials**
Upload this file to the matching `LoadAudio` node:
`LoadAudio` node 11 · `fish_audio_example.mp3`
## Account and pricing
The nodes require signing in to a ComfyUI account and are billed per use. Drama 3 costs about $14.30 per 1M bytes of text during the launch offer, then $35.75 from November 13, 2026. See [Partner Node pricing](/tutorials/partner-nodes/pricing) for the credit rates and the offer window.
## Get started
1. Update ComfyUI to 0.39.3 or newer
2. Open one of the Drama 3 templates above, or add the Fish Audio nodes from `partner/audio/Fish Audio`
3. Select `drama-3` in the Text to Speech node and connect a voice through the Voice Selector or the Instant Voice Clone
4. Write the script with [directions and word tags](#directing-a-performance), then run it
This documentation is built and hosted on [Mintlify](https://mintlify.com), a developer documentation platform.