Directions in brackets
Start a line with how to say it, such as [barely holding it together, forcing a smile], and Drama 3 acts it. Plain words work; there is no fixed list.
Fish Audio Drama 3 AdAnt · by Fish Audio
Drama 3 is Fish Audio's voice acting model: write a direction in brackets before the line, and it plays the emotion.

Play a sample, then use the same setting for your own line.
“[barely holding it together, forcing a smile] I'm fine. Really. [shaky breath] [voice breaking] I just thought you'd at least call.”
“[holding back tears] You promised you'd be there. You promised.”
“[furious, voice rising] You promised you'd be there. You promised.”
“[terrified whisper] You promised you'd be there. You promised.”
“[warm and smiling, like telling a friend] Okay, this is the one I keep coming back to.”
We ran this script once on each model, with no retries and no edits: a warm, friendly woman in her late twenties, casual like a creator talking to a friend: described on MiniMax, and the preset voices Sarah on Fish Audio and Jessica on ElevenLabs. Whether it sounds like a person talking rather than reading, and where it pauses.
Okay, real talk. I've tried every serum on my feed, and this is the only one I've ever finished twice. Two drops, every morning. My skin just looks awake now.
| Model | Made by | Modes | Longest text | Inputs | Test run | Test cost | Best for |
|---|---|---|---|---|---|---|---|
| Fish Audio Drama 3 | Fish Audio | Text to speech, multi-speaker dialogue | — | Script with bracketed directions | 3 s | 0.01 credits | short dramas, audiobooks and emotional ad reads |
| MiniMax Speech 2.8 | MiniMax | Text to speech | — | Script, voice description or a saved voice | 7 s | 0.08 credits | ad voiceovers, UGC and narration |
| Fish Audio S2 Pro | Fish Audio | Text to speech, multi-speaker dialogue | — | Script with bracketed directions | 2 s | 0.02 credits | even, natural reads and narration |
| Eleven v4 | ElevenLabs | Text to speech | 10,000 characters (8,000 on AdAnt) | Script with audio tags | 4 s | 0.07 credits | expressive ad reads, characters and narration |
| Eleven v4 Turbo | ElevenLabs | Text to speech | — | Script with audio tags | 2 s | 0.04 credits | drafts, long scripts and fast turnaround |
Test run and cost come from the same-script run above: the seconds until the finished file came back, and the credits for its characters. Longer scripts cost more; the try-it panel shows the price before you run.
Start a line with how to say it, such as [barely holding it together, forcing a smile], and Drama 3 acts it. Plain words work; there is no fixed list.
Add a new direction or a sound tag like [shaky breath] partway through, and the read turns there. Our example goes from a forced smile to a breaking voice.
Drama 3 can voice a dialogue between several characters in one run, each with its own voice, which suits audio dramas and game scenes. AdAnt's API takes two to four Fish voices for it.
It lists at $0.01 per 1,000 characters on OpenRouter (as of November 2026), so a full short-drama episode costs cents, not dollars.
[who they are and how they feel, e.g. barely holding it together, forcing a smile] First line. [a sound, e.g. shaky breath] [the next feeling, e.g. voice breaking] Second line, with <emphasis>one word</emphasis> stressed.
Fish Audio Drama 3 is a text-to-speech model from Fish Audio, released on October 13, 2026, built for voice acting rather than plain reading. You direct each line in brackets and it plays the emotion, in 83 languages, with open weights. AdAnt runs it with a fixed preset voice, so you can try it here without an account at Fish Audio.
Tears, anger and whispers for every scene of a vertical series.
A testimonial that sounds moved, not read off a card.
Narration and character voices directed line by line.
One emotional scene on Drama 3, Eleven v4 and v4 Turbo, with the price of each.
Ten stock short-drama roles, from the cold CEO to the scheming rival, each with a sample.
Narration voices for audiobooks, documentaries and faceless videos.
Drama 3 is Fish Audio's voice acting model, released October 13, 2026. You write how to say a line in brackets and it acts it, in 83 languages. Our 11-second test voiceover was ready in under 3 seconds for 0.01 credits.
Running this model uses paid credits from an AdAnt plan. Creating an account is free. A short voice line costs well under 1 credit.
On OpenRouter it lists at $0.01 per 1,000 characters (as of November 2026). Fish Audio's own API launched it at $10 per million characters, 80% off its $50 list price, until November 12, 2026. On AdAnt a short line costs 0.01 credits.
Yes: Fish Audio's own API, OpenRouter as fish-audio/drama-3, or AdAnt's media API or the command adant-cli media audio generate --model fish-drama-3, paid in AdAnt credits with no Fish Audio key.
On Fish Audio, yes: Fish Audio offers voice cloning in its own app. AdAnt does not clone voices: it runs Drama 3 with Fish Audio's preset voices only.
VideoGoogle · fast drafts and talking clips with sound
VideoByteDance Seed · longer story clips and complex scenes with many references
VideoByteDance Seed · UGC ads and product clips with references
VideoByteDance Seed · testing many hooks and ideas cheaply
VideoTopview · longer single takes and calm, natural clips
MotionKuaishou (Kling AI) · putting a consistent character into a proven performance
ImageGoogle DeepMind · photoreal product shots and keeping a product on brand
ImageGoogle DeepMind · fast social images, thumbnails and drafts
AdAnt's agents research your market, write the script, pick the right model for each shot and make the video in one chat.
Start with AdAnt