How to use AI Narrator
Paste a script written for the ear
Paste the text to be read aloud into "Paste the script". Write short sentences with one idea each, e.g. "Today I will cover three things." One run takes up to 3,000 characters, about three minutes of narration.
Check the narrator settings
The settings start on a steady narrator: normal speed and Steady expression for an even read. Pick another voice under Voice, set Language to the language of the script (it starts on the language of this page) and choose Slow for dense material. A chip under "Style preset" replaces your text with a sample script, useful for comparing voices.
Listen, fix the lines, run again
Press "Generate narration" and listen for rushed passages and misread names. Put a period where you want a pause, spell a name the way it sounds, then generate again. For a whisper or a sigh, add a tag like [whispers] before the words and switch Expression to Natural. Download the audio, or use it with lip sync so a presenter says it on video.
What a real job costs here
Runs on this page cost 15 credits. At the best pack rate that is about $0.1596.
| Pack | Credits | Thousand characters with ElevenLabs v3 |
|---|---|---|
| $9 | 750 | 50 |
| $19 | 1,650 | 110 |
| $39 | 3,500 | 233 |
| $79 | 7,300 | 486 |
| $149 | 14,000 | 933 |
What this model is good at and where it stops
ElevenLabs v3: Voiceovers, ads and narration that need emotion and many languages. Expressive delivery steered by audio tags. Watch out for: needs more script work than earlier ElevenLabs models.
Narrator settings out of the box
The narrator starts at normal speed with Steady expression: an even, predictable read for documentaries, lessons and explainers. For a story that needs more feeling, switch Expression to Natural or Expressive. The voices and languages are the same as in the AI Voice Generator.
Write for the ear
Short sentences. One idea each. If a line is hard to read aloud, it will be hard to listen to.
Pacing is the tell
A narrator that reads at one speed sounds synthetic no matter how good the voice is. Write short sentences, put periods where you would pause and use Slow for dense material. Commas do less than people expect.
Then put it on camera
Pair the narration with lip sync to make a presenter say it. Or drop it under generated footage as a voiceover.
Frequently asked questions
Is this the same as the voice generator?
Same voices and languages, with narration defaults: a steadier read at normal speed. The voice generator starts on a more expressive setting for ads and voiceovers.
How long can the script be?
Up to 3,000 characters per run, about three minutes of narration. It costs 15 credits per 1,000 characters, and each started 1,000 counts in full. Split a long text into sections, so a weak passage can be redone on its own.
Which languages can it read?
14: English, Polish, German, Spanish, French, Italian, Portuguese, Dutch, Swedish, Japanese, Korean, Chinese, Indonesian and Hindi. Write the script in the language you pick, because the setting does not translate.
Can I use it commercially?
Yes on paid packs, under the UGESI terms.
