Read your script.Talk naturally.It follows your voice,on your device.
Say it reads the headline with your voice (a 41 MB model downloads once). Paced runs it with no download.
Free. No account. Opens offline once loaded. Your script and voice stay on your device.
Online
The whole studio, on your device.
Script and rundown on the left, the stage in the middle, the inspector on the right. On a phone it is one pane at a time with a floating dock.
- 0:00
Paste
Bring a script from Google Docs or Word, or open a .txt or .md file.
- 0:42
Read aloud
The prompter finds your place from the words you say and keeps pace.
- 1:04
Record the take
Camera behind the text if you want it. The take stays on your device.
Three ways to read.
Read next to your own camera
Open VocalScroll on a second screen, a tablet or a phone and record with your camera, OBS or any app. The script follows your voice; nothing is recorded unless you ask.
Record inside VocalScroll
Put the camera behind the text, record video or audio, and keep the take on your device. Voice cleanup and captions work on the take.
Float over other apps
A small Picture-in-Picture prompter stays above other apps where the browser supports it. Paced and manual scrolling are reliable there; voice following depends on the other app leaving the microphone free.
Try it now. No install, no sign-up.
This is the real engine, running in this tab. Press the microphone and read the text out loud, or use paced scrolling, which needs no download.
Welcome to VocalScroll. Read this paragraph out loud and watch the text follow your voice. Pause, and it waits for you. Speed up, and it keeps pace.
Everything here runs on your device. Once the speech model has loaded you can turn off your Wi-Fi and keep reading. When you are ready, take this script into the studio and record a take.
Press the microphone to start.
Edit the text
Add a # line for a section, [pause 2s] for a cue, [color=yellow]word[/color] for colour.
Free. No account. Your words stay on your device.
- 23languages, each with its own speech model
- 41 MBfor the English model, downloaded once
- 0scripts or recordings sent to a server
Seven things, one studio.
The matcher finds your placefrom the last few words.
Follows your voice
The matcher finds your place from the last few words, so it forgives a wrong word or a skipped line.
Camera behind the text
See yourself behind the script, with the text kept readable against the picture.
Sections and rundown
Mark sections with a # line. The rundown lists each with its run time and jumps to it.
Hi, I am Sam. [pause 2s]
steady pacing not scrollingColour tags and cues
Colour a word, or add a stage direction like [pause 2s] that the prompter shows but never waits for you to say.
Captions
Export SRT or VTT files timed from the words you actually spoke.
Floating prompter
Keep a small prompter window above your other apps while you record elsewhere, where your browser supports Picture-in-Picture.
Clicker and pedal keys
Map any key to start, pause and move through sections, so a presentation clicker or page-turner pedal works.
Sections, colour and cues
Write plain text and add a little syntax. The prompter and the rundown are built from it. This example was parsed by the same code the studio uses.
# Cold open
Hi, I'm Sam. [pause 1s] Today: one trick for [color=yellow]steady pacing[/color].
# The trick
Read at the speed you [color=yellow]speak[/color], not the speed you scroll. [smile]
# Call to action
If this helped, try it on your next take.Hi, I'm Sam. pause 1s Today: one trick for steady pacing.
Read at the speed you speak, not the speed you scroll. smile
If this helped, try it on your next take.
Click a section, or drag along the bar, to move the playhead.
From script to captions.
Bring your script
Paste from Google Docs or Word, or open a .txt or .md file. You see an import preview before anything is saved.
Mark sections and cues
Add # headings, colour tags and [cues]. The rundown shows run time per section.
Record a take
Choose video or audio, turn on the camera behind the text if you want it, and press Go. Takes stay on your device.
Clean up the voice
Background noise removal and a mastering preset apply when you play a take back or export it. The original recording is kept as saved. All of it runs on your device.
Export captions
Download SRT or VTT for the take, or the media file itself, when you decide to.
Where your words go.
The privacy claim is a list you can check. Open your browser's network panel while you read: apart from this site's own files, the only requests are the model download and a page-view count.
| What | Where it lives | What leaves the device |
|---|---|---|
| Your script | Saved in your browser's storage, on this device. | Nothing |
| Your voice | Read in memory by a speech model running in WebAssembly, then discarded. | Nothing |
| Takes | Stored in your browser's storage, on this device. | Nothing, until you export a file yourself |
| Speech model | Downloaded once per language from models.vocalscroll.com and kept on your device. | A request for the model file |
| Page views | Counted by Pirsch Analytics, which uses no cookies. | A page view, with the browser details and IP address any request carries. No scripts, voice or recordings |
Fonts and icons are served from this site. No ads, no cookie banner, no account. The full list is in the privacy policy.
What it does
VocalScroll is a voice-activated teleprompter. Paste a script into the editor, press start, and read aloud. Speech recognition on your device follows the words you say and moves the script with your pace. A pause holds the scroll. A faster pace moves it forward. The studio can show a camera preview and record a clip that stays on the device. Scripts in the vault stay on the device as well. The app runs in the browser, and you can install it as a web app on a computer or a phone. It reads scripts in 23 languages, including English, Spanish, French, German, Italian, Portuguese, Chinese, Russian, Japanese, Korean, Hindi, Turkish, Vietnamese, Dutch, Catalan, Persian, Ukrainian, Kazakh, Swedish, Esperanto, Czech, Polish, and Uzbek. Open the studio from this page to load a script and start speaking.
Read in your language
23 languages, each with its own on-device speech model. Pick one to put a sample in the demo, see the model size, and open its page.
English · 41 MB download, once
Welcome to VocalScroll. Read this paragraph out loud and watch the text follow your voice. Pause, and it waits for you. Speed up, and it keeps pace.
All 23 language pages
Who reads with it.
- Video creatorsRead to the lens without cutting between takes.
- PodcastersKeep outlines, sponsor reads and intros in view.
- Teachers and studentsLectures, presentations and recorded assignments.
- Speakers and preachersRehearse at your own pace before the room.
- Sales and interviewsHold eye contact while you stay on message.
- Live streamersTalking points on screen while you go live.
How it compares.
| Feature | VocalScroll | voice-scroll.com |
|---|---|---|
| Use without installing | Yes, in any browser | App store install only |
| Free script length | No cap set by the app | 300 characters |
Checked on 2026-10-09 against voice-scroll.com's public pages. Limits change; tell us if this is out of date.
Frequently asked questions
How does a voice-activated teleprompter work?
VocalScroll listens through your microphone as you read your script out loud. It recognizes what you are saying in real time and automatically scrolls to keep up with your pace. When you pause to take a breath or a sip of water, the scrolling pauses too. When you speed up, it speeds up. It is like having someone follow along with their finger, but hands-free.
Is it free? Is there a script limit?
It is free. VocalScroll sets no cap on script length, and there is no account, no trial and no watermark. Your browser's storage is the practical limit.
Do I have to install anything?
No. Open VocalScroll in a current browser on a computer, phone or tablet and use it straight away. Installing it as a web app is optional.
Can I use VocalScroll for YouTube videos and podcasts?
Yes. You can record yourself on video while reading your script, capture audio only for podcast episodes, or just use it as a scrolling text reader while you record with your own software like OBS, Riverside, or Descript.
Where do my scripts and recordings go?
Nowhere. Your browser's storage keeps your scripts and takes on your device. VocalScroll has no accounts and no server that stores them. Export a take when you want to share it.
Does it work without internet?
Mostly yes. VocalScroll processes your voice on your device. The first time you use a language it downloads that language's speech model and keeps it on your device. After that you can disconnect from the internet and keep reading, which helps in studios and on location.
What languages does it support?
VocalScroll supports 23 languages including English, Spanish, French, German, Italian, Portuguese, Chinese, Russian, Japanese, Korean, Hindi, Turkish, Vietnamese, Dutch, Catalan, Persian, Ukrainian, Kazakh, Swedish, Esperanto, Czech, Polish, and Uzbek. Each language has its own speech model.
Can I install it as an app on my computer or phone?
Yes. You can install VocalScroll as a desktop app on your PC or Mac directly from Chrome or Edge. On a phone, use your browser's install option, or Add to Home Screen on an iPhone or iPad. Once installed, you can launch it like any other app, and it opens without a connection. A language's speech model downloads the first time you use it, so that first use needs the network.
Does it work behind teleprompter glass?
Yes. The prompter can show normal, mirrored (left to right), flipped (upside down) or rotated 180 degrees text, so it reads correctly in a beam-splitter rig. Pick the mode in Settings or cycle it from the Studio.
Can I read from VocalScroll while another app records?
Yes, with a limit. The floating prompter keeps a small window above other apps where your browser supports Picture-in-Picture. Paced and manual scrolling work there. Voice following can stop if the other app takes the microphone, and VocalScroll says so when it happens.
Ready to read your first take?
Open VocalScroll, paste your script and start talking. No sign-up, no subscription.