File formats
Ten formats, from PDF and EPUB to comics, in the browser and in the app.
A scanned page is a picture of letters, so SpeakCove recognizes the text first and then reads it to you. On this website that works for scanned PDFs; in the app, you point the camera at paper.
(a) A PDF without a text layer goes through text recognition in your browser, line by line, and is then read aloud a sentence at a time. (b) In the app for iPhone and iPad, the camera scans the paper and the text is recognized on the device.
WebPick the language of the text, then open the scanned PDF.AppTap + and Scan document. Photograph the pages, put them in order, then save.
WebEvery page goes through text recognition in your browser: Recognizing text… page 3 of 12AppYour phone or tablet recognizes the pages on the device. Nothing is uploaded.
The voice reads the recognized text with the spoken sentence highlighted on the page. Tap any sentence to go there.
Scans, then speech
Many PDFs are not text at all. A chapter run through the office copier, an article saved from an old journal, a letter someone photographed: each page is a picture of letters, and an ordinary reader finds nothing in it to read. SpeakCove recognizes the letters first, then reads them aloud like any other page.
On the web this happens on its own. When you open a PDF and none of its pages carries text, the reader runs text recognition right in the page, one page after another, and counts them as it goes. Then press play. The highlight moves over the scanned page itself, because every recognized line is laid back over the picture where it was found, and you can tap a line to jump there.
Recognition reads in the language you pick, so set it before you open the file; change it later and the pages are recognized again. All 31 languages work. Two limits are worth knowing. Every page is recognized before reading starts, so a long scan takes a while. And the reader only recognizes PDFs that are scanned throughout: if some pages already carry text, it reads those and skips the ones that are only pictures.
Paper on your desk is a job for the app, on iPhone and iPad. Tap + and Scan document, and the camera finds the edges of the page. Scan as many pages as you need, turn any that came out sideways, then save them to your library as a PDF. The phone recognizes the text itself, with Apple’s text recognition on iPhone and iPad, so nothing is uploaded. On iPhone and iPad the app picks the voice language from the text it finds. The first pages are recognized when you save, the rest while it reads.
The app also checks each page on its own. A PDF that mixes typed and scanned pages reads from start to finish, and a page whose text comes out as garbled symbols is recognized from the picture instead.
Yes. Open it in the reader on this website. If none of its pages has text, the reader recognizes the text in your browser first, then reads it with the spoken sentence highlighted.
In the app, yes: tap + and Scan document and point the camera at the page. The web reader takes files, so a photo needs to be a PDF first.
No. On the web, recognition runs in your browser tab and only the recognition engine and language data download. In the app it runs on your phone or tablet.
Very faint or blurred scans can come back empty. The web reader then says “No readable text found in this PDF.” Try a sharper scan, or scan the paper with the app.
Open it in the free reader. No sign-up, nothing to install. For paper on your desk, scan it with the app.
Ten formats, from PDF and EPUB to comics, in the browser and in the app.
Every voice reads all 31 languages, from English and Español to 日本語.