Audio model · from 1 credit
Generate a voice-over with Fish Audio Drama 3
Fish Audio Drama 3 is the voice model from Fish Audio, released in preview on October 13, 2026, built to act rather than read. Direction goes straight into the text: a bracket tag such as [whispering], [laughing] or [shouting] before a word, or a note in parentheses at the start of a sentence, such as (in a trembling voice). The model performs it without saying it. It can also voice a dialogue between two characters with different voices in the same track, which makes it the tool for audio fiction and character voiceovers.
A voice that acts instead of reading
Fish Audio Drama 3 was released in preview on October 13, 2026. Where most voice models read a text evenly, Drama 3 takes direction like an actor: you write the performance into the text and it plays it. A bracket tag before a word, such as [whispering], [laughing] or [shouting], changes how the next words are said; a note in parentheses at the start of a sentence, such as (in a trembling, terrified voice), sets the tone of the whole line.
The directions are never spoken. In our tests on October 15, 2026, in French as in English, the whispered part came out unvoiced and quiet, the shouted part loud and high, and the "terrified" line clearly higher than the same sentence read without direction.
Two characters in one track
Drama 3 can also play a two-person scene in a single track. Start each line with <speaker:0> or <speaker:1>: the model switches voice at every change of character, without reading the tags. It is the simplest way to produce an audio fiction scene, a sketch or a dialogue ad without editing two tracks by hand.
The model offers no voice list to pick from: it uses its default voice and shapes it with your directions. The price follows the length of the text, directions included, and shows before you launch. For a specific voice you want to reuse, the cloning engines of the Voice-over node remain the right door.
Fish Audio Drama 3 next to its closest siblings
Figures pulled from the same registry as billing: when a price or a cap changes, this table changes with it.
| Fish Audio Drama 3 | Fish Audio S1 | Fish Audio S2 Pro | |
|---|---|---|---|
| Base price | 2 credits / 1,000 characters | 3 credits / 1,000 characters | 3 credits / 1,000 characters |
| Voices | – | – | – |
How to use it on Imaginode
- 1Add a Voice-over node from the toolbar.
- 2Write or wire in the text to read, pick a voice: the price follows the text length.
- 3Generate, listen in the node and download the MP3.
Frequently asked questions
Do the directions have to be written in English?
No. Fish Audio says directions are understood in several languages, and our tests in French worked: whisper, shout and trembling voice directions were performed without being read aloud.
How does it differ from Fish Audio S2.1 Pro?
S2.1 Pro is the most natural voice for a calm read, across 80 languages. Drama 3 is built for performance: emotions, voice effects and dialogue, for slightly less per character.
How much does a voice-over generated with Fish Audio Drama 3 cost?
2 credits per 1,000 characters, about $0.02 (1 credit = $0.012). The exact cost shows before generating, and the MP3 downloads straight from the node.
Can I try Fish Audio Drama 3 for free?
Yes: signing up grants trial credits, no credit card required. Enough to test Fish Audio Drama 3 in real conditions before deciding on a plan or a top-up.
Do I need to install anything to use Fish Audio Drama 3?
No: Fish Audio Drama 3 runs server-side, in the browser, mobile included. Imaginode's node canvas wires its renders to the other models (references, camera, styles) with nothing to install.
Try Fish Audio Drama 3 on a real canvas
126 models, one account, credits that work everywhere. The price always shows before you generate.
Get started
ElevenLabs v3
MiniMax Speech 2.8 HD