Drama 3 · directed speech, not just text to speech
Write the script, describe how each line should be performed, and generate an acted MP3. Directions such as [whispering] or [short pause] sit inside the line exactly where you want them.
Each clip below is a generation of the exact script shown next to it. Open the full script, copy it, or send it to the studio and direct your own version.
[barely holding it together] I'm fine. Really. [shaky breath] Go on without me. I'll catch up. [pause] I just need a minute.
[low, measured, like a documentary narrator] The city went quiet at 3 a.m. [short pause] No sirens. No traffic. <slow>Just the rain.</slow>
[amused, taking his time] You came all this way. [soft laugh] For a file. [harder, dropping the warmth] Then let me show you what is in it.
[smooth, close to the microphone] You're listening to the midnight hour. [sigh] If you're awake with me, <emphasis>you're not alone</emphasis>.
[clear and friendly] Welcome aboard. Today we cover the basics. [short pause] [firmer] And then we cover the mistakes that cost money.
[hushed, telling a secret] Once, in a village that no longer exists, [pause] a girl traded her name for a lantern. <whisper>This is her story.</whisper>
[professional, polite] I understand. [short pause] [slower, holding it together] I have explained that twice. [exhale] Let us try once more.
[gentle, unhurried] She kept the letter for forty years. [pause] [almost a whisper] She never read it again, but she never threw it away either.
What this studio accepts, what it returns, and how directions work.
Drama 3 reads directions as language, not as settings. Four habits cover most of what you will need.
Instead of picking from a fixed list of emotions, describe the intent: who is speaking, to whom, and how hard they are trying. The direction can be a short phrase or a full sentence.
[barely holding it together] I'm fine. Really.
A sigh, a laugh, a breath, or a pause can be placed on its own. Put it where it belongs and it will not leak into the rest of the line.
[sigh] Let's try that again. [short pause] Slower this time.
Paired tags apply only to the words inside them. Wrap the phrase you want changed and leave the rest of the sentence as written.
I'm fine. Really. <whisper>Go on without me.</whisper>
Directions are not limited to English. If the performance is easier to describe in your own language, write it there; the delivered line stays in the language of your script.
[像是在强忍着眼泪，却努力笑着] I'm fine. Go on without me.
Paste a line, a monologue, or a short scene. Keep the wording plain; the directions do the directing.
Use a chip or type your own brackets, then read the line back. Small wording changes usually change the performance.
The studio generates the script and returns one MP3 you can preview, replay, and download.
One script becomes one performance. Line by line, that covers a lot of ground.
Direct the quiet passages and the tense ones differently instead of reading the whole chapter at one level.
Generate a character's line with the exact attitude you need, then regenerate until the take lands.
Set a warm opening, a firmer middle, and a clear close inside the same short script.
Keep the instructional lines plain and the important warnings pointed.
Record each role's line as its own file and cut them together in your editor.
Test a line five ways before deciding how a script should sound.
Most text-to-speech tools let you pick a mood. This one lets you write the performance.
Direction chips insert the bracket at your cursor, so the script and the performance notes stay in one place.
A whisper or a slowdown can be limited to a phrase, leaving the rest of the line at full voice.
Write the direction in the language you think in, while the script itself stays in the language you are producing.
Every example on this page shows the full script beside a real generation, so you can borrow the structure before you write your own.
Drama 3 is a text-to-speech model from Fish Audio built for directed performances. This studio wraps it in a script editor: you write the line, add directions, and download the resulting MP3.
Wrap that word in a paired tag, for example <whisper>Go on without me.</whisper>. The effect applies to the words inside the tag, and the sentence around them is spoken normally.
Yes. Directions are read as language, so you can write them in your own language while the script itself stays in the language you are producing.
Up to 2000 characters per generation, counting the directions inside the brackets.
Not in a single request. Generate each character's line as its own file and cut them together, which also lets you direct every line separately.
Yes. Drama 3 is a preview model and Fish Audio notes that results vary between takes, so generating a few and keeping the best one is normal.
No. This studio generates speech from your script and directions; voice cloning is not offered.
Paste your script, add the directions that carry the performance, and download the MP3.
