How to Make Your First Text-to-Speech Voiceover in ElevenLabs

ElevenLabs converts written text into spoken audio with voices that sound close to a real person. The fastest way to learn the tool is to generate one short clip end to end. This guide walks you from a blank box to a downloaded MP3 in a few minutes, using only the free tier.

What you need

• A free ElevenLabs account (sign up at elevenlabs.io)
• A short script, ideally one or two sentences for your first try
• A modern browser; nothing to install

Step 1: Open the Text to Speech studio

After logging in, open the left sidebar and click Text to Speech. You land on a single large text box with a voice picker above it and a generate button at the bottom. This is where every voiceover starts.

Step 2: Pick a voice

Click the voice dropdown and choose one of the prebuilt voices such as Rachel or Adam. Each name has a small play button so you can hear a sample before committing. For a first test, any default voice works fine.

Step 3: Choose a model

Set the model to Eleven Multilingual v2 for the best quality on most languages, or Eleven Flash v2.5 if you want the fastest, cheapest output. Quality models cost more characters per generation, so keep test clips short.

Step 4: Paste your text and generate

Type or paste your script into the box, then click Generate speech. After a few seconds an audio player appears below the box. Press play to listen. Each generation counts against your monthly character quota, so the free tier is plenty for learning.

Step 5: Download the audio

Use the download icon on the player to save the clip. On the free plan you get an MP3 at a standard bitrate, which is good enough for most online video. Paid plans unlock higher quality exports including WAV.

Result: a single sentence like a video intro becomes a clean MP3 you can drop straight onto a timeline in CapCut, Premiere, or DaVinci Resolve, with no microphone required.