Rescroll

A free AI voice over generator, for your videos.

Paste your video script, choose a narrator from 26 natural AI voices, and set the speed and the pauses. Download the voice-over as an MP3 for YouTube, TikTok, Reels or an ad. Free, no sign-up, no watermark, and nothing is uploaded.

  • 100% free
  • No sign-up
  • No watermark
  • MP3 or WAV download
  • Nothing uploaded

Loading the tool…

It runs in your browser. Nothing is uploaded.

How it works

How to turn text into speech, free, in three steps.

  1. 1

    Type or paste your text

    Up to 25,000 characters at a time, about half an hour of speech. A blank line starts a new paragraph, and a note like [pause 2s] adds a pause there.

  2. 2

    Choose a voice and press Play

    Pick one of 26 natural voices, American or British, and set the speed. It starts speaking after the first sentence and lights up each sentence as it reads.

  3. 3

    Download the MP3

    When it sounds right, download it as an MP3, or as a WAV for editing. No watermark, no sign-up, and the file is yours to use.

Why this one

Free, with nothing held back, and nothing uploaded.

Natural AI voices, free

26 voices from Kokoro, an open-source AI voice model: 18 American and 8 British, women's and men's. Not the robotic voices built into your computer.

Starts in a moment, however long

Long texts are read a sentence at a time: playback starts as soon as the first is ready, the rest are made while you listen, and the sentence being read lights up.

Pauses and pronunciations

Set the pause between paragraphs, or type [pause], [pause 2s] or [long pause] where you want one. Teach it a name it says wrong: respell it, or give its sounds.

Private, and no limits of ours

The voice runs on your device, so nothing you type is uploaded, and there's no daily or weekly character allowance: make as much audio as you like.

Your text is read by a voice model that runs in your own browser, and the MP3 is made there too. Nothing you type is sent anywhere: the only download is the voice model itself, the first time.

Making a voice-over from your script

A voice-over lives or dies on its pacing. Set the speed (a touch slower than a chat reads more clearly over pictures), the pause between paragraphs for each new shot, and type [pause] or [beat] where a point should land. The time it will take shows as you type.

This page opens on a short hook in Michael, a calm American narrator, at 0.95× speed. Try 26 voices in all, American and British: press a voice while it plays and the same sentence carries on in the new one.

Prefer your own voice? Take the script to the free teleprompter and record yourself reading it. Either way, Rescroll's AI video editor can build the video around the voice-over: captions, B-roll and graphics, described in words.

Rescroll

Got the voice-over? Make the video around it.

Rescroll's AI video editor edits from what you type: add the B-roll, the captions and the motion graphics to your voice-over by describing them, then export the finished video. It's paid, with a free plan to start on.

  • Captions timed to every word
  • Graphics described, not keyframed
  • Cuts, zooms and a color grade by asking

Questions

The text to speech tool: your questions, answered.

Can I use an AI voice over on YouTube?

Yes. The voice model, Kokoro, is released under the Apache 2.0 licence, which allows commercial use, and we add no terms of our own, so the MP3 is yours to use on YouTube, including monetized videos. YouTube asks you to disclose realistic synthetic content in some cases: check its current rules for your video.

How long will my voice-over be?

The page shows the expected length as you type, at the speed you set and with your pauses, and the exact length once it's made. As a guide, a natural AI voice at 1× reads about 14 to 15 characters a second, so 1,000 characters is a little over a minute.

Is this text to speech really free?

Yes. There's no price, no trial, no account and no daily or weekly character limit. Each go takes up to 25,000 characters, and you can do as many as you like. The voice runs on your own device, so it costs us nothing to run, and the MP3s have no watermark.

Is my text uploaded anywhere?

No. The voice model runs in your browser, on your device, and the MP3 is made there too. The only thing downloaded is the voice model itself, from Hugging Face, the first time; after that it's kept in your browser. You can check in your browser's developer tools: no text is sent while it speaks.

Why does it download something first?

The voice is an AI model that runs on your device rather than on a server: your browser downloads it once and keeps it. It starts with the processor's model, 119 MB, so you hear it sooner; your graphics card's, 352 MB, makes speech many times faster, and you can switch to it with one click. The next time, it starts in a few seconds.

Which browsers and devices does it work on?

Current Chrome, Edge, Safari and Firefox, on a Mac, a Windows PC, a Chromebook or Linux, and on recent phones. It's fastest where the browser can use your graphics card (WebGPU): in Chrome on an Apple M4 Max it made speech about 11 times faster than it plays in our test. On the processor it can be slower than real time, so playback pauses to catch up on long texts; a computer is much faster than a phone.

Can I use the audio in my videos, even commercially?

Yes. Kokoro, the voice model, is released under the Apache 2.0 licence, which allows commercial use, and we add no terms of our own: the MP3 is yours, for YouTube, TikTok, a podcast, an ad or a course.

Which languages does it speak?

English, in American and British accents: 18 American voices and 8 British ones. Other languages aren't supported yet; English text with a few foreign names works, and the pronunciation box fixes any name it says wrong.

How do I add a pause?

Leave a blank line between paragraphs, and set how long the pause there is (none, short, medium or long). Or type a note in square brackets where you want one: [pause] is a second, [pause 2s] two, [short pause] half a second and [long pause] two. Any other note in brackets, like [smile], isn't read aloud.

It says a word wrong. Can I fix it?

Yes. Under Pronunciations, type the word and how to say it: respelled the way it sounds (Rescroll as re-scroll, GIF as jif), or as the model's sounds between slashes, like /kˈOkəɹO/. Every time the word appears, it's said your way.

More free tools for your videos

Silence remover

Cut the pauses and dead air out of any video or recording, automatically. Full quality, no watermark.

Video to text

Transcribe a video or a recording in 99 languages, with timestamps and SRT or VTT captions. Nothing uploaded.

Add captions to video

Auto-captions burned into your video, word by word, in five looks, clear of TikTok, Reels and Shorts buttons. Nothing uploaded.

Translate video

Speech in 99 languages to English subtitles, or SRT and VTT files into 38 more languages. Nothing uploaded.

Merge videos

Join clips into one MP4 in the order you drag them, any sizes, sound kept. No upload.

Video speed changer

Speed up or slow down a video, 0.25× to 4×, with the voice at its own pitch. No upload.

Reverse a video

Play a video backwards, sound reversed or removed, up to 10 minutes. No upload.

Video looper

Repeat a video a number of times or to a length, fast, at full quality. Boomerang too.

Rotate a video

Turn a video 90°, 180° or 270° in seconds, with no re-encoding and nothing lost.

Flip a video

Flip a video horizontally or vertically, in every frame, as an MP4. No upload.

Filler word remover

Find every um and uh in a video or recording, check them in the transcript, and cut them out. Free, no watermark.

Video compressor

Shrink any video to a target size or quality as an MP4 that plays everywhere. No upload.

Video resizer

Change any video to 9:16, 1:1, 4:5 or 16:9 with a blurred fill or a crop. Full quality, no watermark.

Video cropper

Drag a crop box over any video, free-form or 9:16, 4:5, 1:1 or 16:9. The crop's own pixels, no watermark.

Extract audio from video

Save any video's sound as MP3, WAV or M4A, the original copied untouched where possible.

Add music to video

Put music under any video: trim, loop, fade, and it dips under speech by itself. The picture untouched.

Remove audio from video

Take the sound out of any video in seconds, the picture copied untouched. No re-encoding, no watermark.

MOV to MP4

Convert iPhone and Mac MOV videos to MP4 in seconds, losslessly, with nothing uploaded.

MKV to MP4

Remux MKV to MP4 losslessly, with Dolby Digital converted and the sound track yours to choose.

WebM to MP4

Convert WebM recordings to H.264 MP4 on your device: no upload, no install.

Video to GIF

Turn any clip into a sharp, small GIF, with the size shown before it's made. No upload.

Video trimmer

Cut the start and end off any video, frame-exact or lossless with no re-encoding.

Loudness normalizer

Bring any audio or video to −14, −16 or −23 LUFS, peaks held under −1 dBTP. Video untouched.

Apply LUT to video

Grade a video with your .cube LUT or a free look, see it live, and download the MP4. Nothing uploaded.

TikTok safe zone checker

See TikTok's, Reels' and Shorts' buttons and captions over your video, and download 1080×1920 PNG templates.

Auto reframe video

Landscape to vertical with AI face tracking: the crop follows the speaker, smoothly, shot by shot. Nothing uploaded.

Blur faces in video

Every face found and followed automatically, then blurred or pixelated. Keep the ones you choose. Nothing uploaded.

Video background remover

Remove the background behind people in a video: transparent, green screen, colour, blur or image. Nothing uploaded.

Audiogram generator

A clip of your episode as a video: cover art, a moving waveform and word-by-word captions, in four shapes.

MP3 to MP4 converter

Any MP3 as an MP4 with your image on it, the whole file, ready for YouTube. Fast, even for an hour-long episode.

Bleep out swear words

Find every swear word in a video or recording, check them in the transcript, and bleep them out. Free, no watermark.

Remove background noise

Take the fan, hum, traffic and wind out from under the voice in a video or recording. Hear it before and after. Free.

Teleprompter

Read your script to camera: scrolling at your pace or your voice's, mirrored for glass, and recorded as an MP4.

Speaking time calculator

Paste a script to see how long it takes to say, and how many words fit 30 seconds, a minute or three.

Mic test

A live level meter with a verdict: too quiet, good, clipping or a noisy room, with fixes, and a 5-second playback.

Webcam test

See the resolution and measured frame rate your webcam really delivers, check your light and framing, and take a snapshot.

Online voice recorder

Record your voice, pause, trim and save as MP3, WAV or M4A, kept safe through a reload.

Screen Recorder

Record your screen with your voice and your camera in a bubble, then trim it and save an MP4. No time limit.

Webcam Recorder

Record yourself with your webcam and microphone, pause, trim and save an MP4. No time limit.

YouTube thumbnail size checker

YouTube's thumbnail size and rules, your thumbnail in its feeds dark and light, and a JPG made to fit.

Aspect ratio calculator

Width and height to ratio, a ratio to the missing side, every platform's size, and whether one frame fits another.