Read PDF Aloud (Text to Speech)
Listen to any PDF with your device’s voices — sentence by sentence, nothing uploaded.
You stopped here last time.
Voice and reading options
Reading options
All pages.
Text being read
Click or tap a sentence to read from there.
Space plays or pauses, left and right arrow keys go to the previous or next sentence.
About the Read PDF Aloud (Text to Speech)
Turn a PDF into something you can listen to. The page text is put back into reading order — columns one after another, words split by a hyphen at the end of a line joined again — and running headers, footers and page numbers are left out, so the voice reads the document, not the page furniture. The sentence being read is highlighted in the text beside the controls, and the word too when the voice reports it.
The speech comes from the voices built into your browser and operating system (the Web Speech API). Voices that run on your device are chosen first; voices that send the text to an online speech service are labelled with who receives it, and are never chosen for you. Your PDF is opened on this device and never uploaded.
How to use it
- Press Choose a PDF (or drop it on the box). A password-protected PDF asks for its password, which is used only on this device.
- Pick a voice, then set the speed and pitch. Voices on this device are listed first; online voices are marked.
- Open Reading options to read only some pages (From page / To page) and to choose what is skipped: headers, footers and page numbers, numbered references like [12], and full web addresses.
- Press Play. Click or tap any sentence to read from there; Previous and Next move one sentence at a time, and the Space bar pauses when the text has focus.
- Use Stop after for a sleep timer, and Copy text or Download .txt to keep the text exactly as it is read.
Examples
A 40-page annual report whose pages all start with “ACME Ltd — Annual Report” and end with “Page 7 of 40”
Neither line is read on any page: after a few pages the repeating header and the page numbers are recognised and left out, and the number of skipped lines is shown per page.
A research paper set in two columns, with references like [12] and web links
The left column is read before the right one. With “Skip numbered references” on, “as shown [12], [14–16]” is read as “as shown”; https://doi.org/10.1000/xyz is read as “doi.org”.
From page 31 to page 48, speed 0.9×, Stop after: 20 minutes
Reading starts at page 31, stops at the end of page 48 or after 20 minutes — whichever comes first — and remembers the sentence it reached for next time.
Common uses
- Listening to reports, contracts, papers and e-books while you rest your eyes or do something else.
- Proof-listening your own writing: errors you read past are easy to hear.
- Following a document with the sentence highlighted, which helps with dyslexia, low vision or reading in a second language.
- Revising study notes and textbook chapters by ear.
What is read, and what is skipped
PDFs store text as positioned fragments, so the reading order is rebuilt from the layout: lines are grouped into paragraphs, and columns are read one after another. A line at the very top or bottom of a page (the top or bottom 9 %) that is not larger than the body text counts as a header or footer when it repeats on several pages — numbers are ignored when comparing, so “Page 3 of 40” and “Page 4 of 40” are the same line. Bare page numbers are always skipped. Headings (text set clearly larger, or short bold lines) are shown in bold and read as a sentence of their own.
- Skip headers, footers and page numbers — on by default; untick it to hear every line.
- Skip numbered references — leaves out [12], [3, 5] and [7–9].
- Read web addresses as the site name — https://www.example.com/docs?id=3 is read as “example.com”.
- Say page numbers — the voice says “Page 12” when a new page starts.
Voices and your privacy
The voices are the ones your browser and operating system provide through the Web Speech API. A voice that runs on the device keeps the text there. Some browsers add online voices — for example Google’s voices in Chrome and Microsoft’s “Online (Natural)” voices in Edge — which send the sentence being read to that company’s speech service; they are listed separately, marked “online”, and only used when you choose one. Speed (0.5× to 2×) and pitch are passed to the voice; a voice may not support the whole range.
Keyboard
- Space — play or pause (when the text or the player has focus).
- ← / → — previous or next sentence.
- Tab reaches every control; the text area scrolls with the arrow keys and follows the sentence being read.
Limitations
- The audio can’t be downloaded: browsers don’t let a web page record their built-in voices, so the speech is only played.
- Scanned pages are pictures with no text to read. Make them searchable with OCR PDF first, then open the new file here.
- Tables, formulas and text in pictures are read as plain text in reading order, or not at all; footnotes are read where they sit on the page.
- Which voices, languages, speeds and word highlighting are available depends on your device and browser. Phones may pause the speech when the screen locks or the browser goes to the background.
- Some PDFs don’t say which letters their fonts draw (it happens with some Hindi and other complex-script PDFs); those characters can’t be read aloud.
Privacy
Your PDF is opened and read in your browser and is never uploaded. With a voice on this device, the text stays on this device too; if you choose a voice marked “online”, the sentences being read are sent to that voice’s speech service (Google, Microsoft or the browser maker). If “Remember where I stopped” is on, this browser stores the page and sentence you reached for each file, under a fingerprint made from the file’s name, size and first bytes — not the name or the content itself.
Frequently asked questions
Can I save the reading as an MP3 or other audio file?
No. Browsers don’t give web pages a way to record their built-in voices, so the speech can only be played. You can keep the text with Download .txt.
Is my PDF uploaded?
No. The PDF is opened with pdf.js in your browser and the text is spoken by your device’s voices. Only if you pick a voice marked “online” are the sentences sent — to that voice’s speech service, not to MySmartCoPilot.
Why doesn’t it read the header, footer or page number?
Lines that repeat at the top or bottom of the pages are left out on purpose, so they don’t interrupt every page. Untick Skip headers, footers and page numbers in Reading options to hear them.
Why does it say there is no text to read?
The page is most likely a scan — a picture of text. Run the PDF through OCR PDF to add a text layer, then open the searchable copy here.
How do I get more or better voices?
Add them in your system settings and reload this page: on Windows 11 under Settings → Time & language → Speech, on a Mac under System Settings → Accessibility → Spoken Content → System voice, on iPhone and iPad under Settings → Accessibility → Spoken Content → Voices, and on Android under the text-to-speech output settings (usually in Accessibility).
Does it remember where I stopped?
Yes, on this device, when Remember where I stopped is ticked: open the same file again and press Continue to carry on from the sentence you reached.
How is this different from the PDF Reader’s read-aloud?
The PDF Reader reads the page you are looking at. This tool is made for listening: it skips headers, footers and page numbers, shows the text being read so you can start from any sentence, reads a page range, and has pitch, a sleep timer and a time-left estimate.