Script in. Full-cast audio out.
AI Audio Drama Generator
Write a script, cast up to 16 characters, direct the emotion of every line, and get one finished multi-voice MP3. AudioDrama is an AI audio drama maker for podcasts, audiobooks, games and stories.
Sign in with Google and get 2 free minutes. No card needed.
Narrator[calm, low voice] The lighthouse had been dark for forty years. Tonight, a light moved inside it.
Mara[whispering] Did you see that? Someone is up there.
Jonas[nervous laugh] It's the wind, Mara. It's always the wind.
Mara[short pause] [firm] Then why did the door just open?
How it works
Three steps from page to performance
Write or paste a script
One line per turn, in the form
Name: line. Put emotion and delivery in[square brackets]and stress words with<emphasis>.Cast the voices
Characters are detected automatically and given distinct voices. Change any of them.
Generate and download
Get one MP3 of the whole scene, saved to your private library.
Script format
Plain text. Stage directions included.
Turn a script into an audio drama with plain text. Each line starts with the character’s name. Lines without a name are read by the Narrator.
Directions like [whispering] or [short pause] shape the delivery and are not spoken. <emphasis>…</emphasis> stresses a word.
Narrator: [calm, low voice] The lighthouse had been dark for forty years. Tonight, a light moved inside it.
Mara: [whispering] Did you see that? <emphasis>Someone</emphasis> is up there.
Jonas: [nervous laugh] It's the wind, Mara. It's always the wind.
Mara: [short pause] [firm] Then why did the door just open?Emotion per line
Directable emotion per line
Tell the voice how to say each line: whispering, angry, laughing, a short pause. It works like stage directions in a script. Change a performance by editing a word in brackets rather than re-recording.
[whispering]
“Did you see that?”Direction in brackets. Not spoken aloud.
[angry, through gritted teeth]
“Did you see that?”Same words, different delivery.
[laughing]
“Did you see that?”Results vary by direction, so listen and adjust.
Features
Everything a table read needs
Multi-voice text to speech in one take
The whole scene is generated in a single pass, so the conversation flows as one recording instead of clips you stitch together.
MP3 download
Download the finished audio as an MP3 and use it in your editor, podcast host or game engine.
Private library
Every drama you make is saved to your account. Play, download or delete it whenever you like.
Voices
16 curated studio voices
Choose from narrators, young and older voices, and British, Irish and Australian accents. Voices are built in; there is no voice cloning.
Voice names and descriptions are listed in the app. We don’t host pre-made recordings on this page.
Use cases
Made for people who tell stories
- Fiction podcastsDraft full-cast episodes without booking a cast.
- AudiobooksGive each character a distinct voice and let the Narrator carry the prose.
- Game and animation dialogueUse it as an AI voice acting generator for prototypes, placeholder lines and indie projects.
- Short video storiesRecord dialogue for shorts and explainers in minutes.
- Language learningCreate conversations between speakers for listening practice. Voices are English; other languages may work, results vary.
Why a drama maker
AI audio drama maker vs. single-voice TTS
A normal text-to-speech tool reads a page with one voice. To get a conversation you have to generate each character separately, then line up the clips and fix the timing yourself. An AI audio drama maker works from a script instead: it knows who is speaking, gives each character a voice, and reads the whole scene in one pass, so the pauses and turn-taking sound like a conversation. Multi-voice text to speech with per-line directions also means you can change a performance by editing a word in brackets rather than re-recording.
Sample
Try the example script
This is the example script from above. We don’t host a pre-made recording here.
Sign in to hear it — you get 2 free minutes. Open the generator, paste the script and press Generate.
Pricing
Pay once, use whenever
Credits do not expire. Prices are in USD.
Free
$0
2 minutes when you sign up.
60 minutes
$9 one-time
60 minutes of generated audio.
240 minutes
$29 one-time
240 minutes of generated audio.
FAQ
Frequently asked questions
What is an AI audio drama generator?
An AI audio drama generator turns a written script with several characters into finished audio. With AudioDrama you paste a script, give each character a voice, and get one MP3 in which every character speaks in their own voice, with the emotion you directed.
How do directions work?
Put a direction in [square brackets] inside a line, for example [whispering], [angry] or [short pause]. The text-to-speech model reads it as an instruction for how to deliver the line instead of speaking it aloud. Wrap a word in <emphasis>…</emphasis> to stress it. Results vary by direction, so listen and adjust.
How many characters can my script have?
Up to 16 characters in one script, plus an optional Narrator for any line without a name. A single script can be up to 6,000 characters long. Split longer stories into chapters and generate each one separately.
Which languages are supported?
The 16 built-in voices are English voices. The Drama 3 model supports 83 languages according to Fish Audio, so lines and directions written in other languages may work, but results vary and we cannot guarantee accent or pronunciation quality outside English.
Can I use the generated audio commercially?
You keep ownership of the scripts you write, and we do not claim rights to the audio generated from them. You are responsible for having the rights to your script and for following our Terms, including the rules against imitating real people. This is not legal advice; check the Terms for the details.
Can I clone a voice?
No. AudioDrama only offers a fixed set of 16 curated studio voices. You cannot upload voice samples, and using the service to imitate or clone a real person’s voice without their consent is prohibited.
How are credits counted?
Credits are counted by the audio length or the script length, whichever is greater. Script length is 1 second of credit per 15 bytes of script text, including [directions] (about 15 English characters; non-English text uses more bytes per character). A 90-second drama of ordinary English dialogue uses about 90 seconds of credit. While a drama generates, a hold is reserved from your balance and the unused part is returned afterwards; failed generations are fully refunded. Credits are bought once and do not expire.
Can I get a refund?
Credit packs are one-time purchases handled by Creem, our merchant of record. Refunds follow the refund terms in our Terms of Service, Creem’s policy and applicable law. Contact support@hueflo.com if something went wrong.
Which AI model does AudioDrama use?
AudioDrama sends your script to Fish Audio’s Drama 3 text-to-speech model through its API. AudioDrama is an independent product by hueflo.com and is not affiliated with Fish Audio.
Is AudioDrama free?
You can try it free: every new account gets 2 minutes of audio when you sign in with Google. After that, you can buy a credit pack for $9 (60 minutes) or $29 (240 minutes), with no subscription.
Start with 2 free minutes
Powered by Fish Audio’s Drama 3 text-to-speech API. AudioDrama is not affiliated with Fish Audio.