What is screen-to-speech?
Screen-to-speech (also called screenshot-to-speech) is a way of listening to text that a computer cannot select: you frame a part of your screen, the pixels in that frame are turned into text, and the text is read aloud. Because it works on the picture rather than the file, it reaches text inside images, scans, dialogs and remote windows.
The term describes a category, not one product. This page explains what belongs in it, what does not, and where the confusion with screen readers comes from.
How does screen-to-speech work?
In four steps, and on a good tool the whole thing takes under a second. Nothing needs to be a file, and nothing needs to be selectable.
Frame a region
You press a hotkey and drag a frame around the part of the screen you want to hear. That frame is the whole input.
Recognise the text
Text recognition turns the pixels inside the frame into characters. On Windows and macOS this engine is part of the operating system, so it can run on your own machine; on Windows it takes 50-150 ms.
Split it into sentences
The recognised text is cut into sentences, and the language of each sentence is worked out, so a voice can be chosen per sentence rather than per document.
Read it aloud
A voice speaks the sentences, usually with a window that highlights the sentence being spoken so you can follow along or click back.
How is screen-to-speech different from other reading tools?
Four neighbouring categories get mixed up with it. The difference is always the same two questions: what does the tool read, and what do you have to give it?
| Kind of tool | What it reads | What you give it | Examples |
|---|---|---|---|
| Screen-to-speech | Any pixels on your screen, including text inside images | A frame you drag | Speakybara |
| Text-to-speech app | Text you paste, type or select | Selectable text | Web readers, Balabolka |
| Document reader | Files you import: PDF, EPUB, DOCX | A file, and a library to put it in | NaturalReader, ElevenReader |
| Screenshot tool | Nothing. It saves the picture | A frame you drag | Snipping Tool, ShareX, CleanShot X |
| Screen reader | The interface itself: menus, buttons, text fields | Nothing. It drives the whole system | NVDA, Windows Narrator, VoiceOver |
How is it different from a text-to-speech app?
A text-to-speech app needs text it can already select or paste. Screen-to-speech starts one step earlier: it recognises the text first, so a scan, a screenshot or a dialog box works exactly like a web page.
How is it different from a document reader?
A document reader is built around files and a library: you import a PDF or an ebook, it keeps your place, and it often exports audio. Screen-to-speech imports nothing and keeps no library — only a local history you can switch off, never uploaded; it reads the part of the screen you are looking at, right now.
How is it different from a screenshot tool?
A screenshot tool captures the same frame, but the output is an image: you annotate it, save it or share it. Screen-to-speech throws the image away and keeps the text. The two sit next to each other happily.
Is screen-to-speech the same as a screen reader?
No, and the words are worth keeping apart. A screen reader such as NVDA, Windows Narrator or VoiceOver lets people who are blind operate the entire computer by voice; it reads the interface, and it cannot read text that only exists as pixels inside an image. Screen-to-speech is a read-aloud tool for one selected region, and it is a companion to a screen reader rather than a replacement.
When does screen-to-speech help?
Whenever the text you want to hear is visible but not selectable, or when you would rather listen than read. These are the everyday cases.
- A scanned PDF with no text layer: a letter, a contract, an old report.
- An image in an email, or a screenshot a colleague pasted in a chat.
- An error dialog you cannot copy from.
- A dashboard, a chart label or a map.
- A remote desktop or Citrix window, where the text is drawn as pixels.
- A paused video or a slide.
- Any long text at the end of the day, when your eyes have had enough.
- Reading in a second language, where hearing a sentence lands better than reading it.
Which screen-to-speech tools are there on Windows?
A handful, and they split neatly into the ones that copy text and the ones that read it aloud. Only some do both.
| Tool | Copies the text | Reads it aloud | Price |
|---|---|---|---|
| Speakybara | yes | yes, with a read-along window | free |
| Speechify (Windows app) | not stated | yes, screenshot-to-listen since March 2026 | free tier, Premium published in USD |
| PowerToys Text Extractor | yes, Win+Shift+T | no | free |
| Windows Narrator | no | reads the interface, not text inside images | free, part of Windows |
Checked October 2, 2026. On macOS, Speakybara runs as a test version too; the best-known alternative there is TextSniper, which copies on-screen text and can speak it with the system voices. See the full comparison
What is an example of a screen-to-speech app?
Speakybara is one, and it is the app this site is about.
Speakybara is a free app for Windows and Mac that turns any part of your screen into text you can copy or listen to: press a hotkey, drag a frame, done.
Last updated September 9, 2026