How to make an AI voiceover with ElevenLabs on the free plan
Disclosure: some links on this page are affiliate links. If you buy or book through them I may earn a commission, at no extra cost to you. How this site makes money
This walks through turning a short script into a natural-sounding voiceover. It takes a few minutes and works on the free plan.
Before you start: what the free plan allows
The free plan gives a monthly allowance of credits, used up by the characters you generate. On the pricing page, the commercial license is listed from the Starter plan up, not on Free. So the free plan is for trying the tool and for personal projects; if your video earns money or promotes something, plan on a paid tier. Check the current terms for whether free-plan audio must credit ElevenLabs.
Steps
- Create an account and open Text to Speech.
- Paste your script. Keep it short for the first try, around 100 words.
- Pick a voice from the library. Play the previews; choose for clarity at normal speed, not for character.
- Generate, then listen all the way through once.
- Fix problems in the text, not the settings. See the next section.
- Download the audio and drop it into your editor.
Getting it to sound right
- Punctuation is your pacing control. A comma is a short pause, a period a longer one. Break long sentences in two.
- Spell out what should be spoken. Write “twenty twenty-six”, “three point five”, “U S dollars” if the default reading is wrong.
- Names and brands: write them the way they sound if they are mispronounced.
- Regenerate small pieces. If one sentence is off, generate just that sentence again and splice it, which also saves credits.
- Stability setting: higher is more even and neutral; lower is more expressive and less predictable. Start in the middle.
Saving credits
Draft and proofread the script before generating. Generate in paragraphs, not the whole script at once, so a single mistake does not cost a full regeneration. Keep the good takes; you can reuse an intro or outro line across videos.
Disclosure and limits
Do not clone or imitate a real person’s voice without their permission. Some platforms ask you to label realistic synthetic media; a generic narrator over your own footage is usually treated differently from an imitation of a real person, but check the rules of the platform you post on.
When to use something else
If you only need captions, you do not need a voice at all. If you want your own voice, a phone recording in a quiet room with a blanket over your head still beats any model for sounding like you.