Runs in your browser with pdf.js and your device's voices
PDF Listener — hear any PDF read aloud, and follow every word
Add a PDF or paste text and press play. The sentence being read is highlighted, you can click any line to jump there, and headers and page numbers are skipped. English, Hindi, Gujarati and other languages work with the voices on your device.
- PDF not uploaded to our server
- Sentence and word highlight
- Click any line to jump
- Indian languages too
Listen to a PDF
Add a PDF or paste text, pick a voice, and press play. Each sentence is highlighted as it is read.
Drop a PDF or text file here
or browse: a PDF or a .txt file
Or paste text
Text
Click a sentence to read from thereVoice and speed
Your voice and speed are remembered on this device. Shortcut: Space plays and pauses while the text has focus.
Your pages
The page being read is outlined. Use Read from here on any page, or click a page to see it full size.
What this tool does
Your browser reads the PDF aloud. You follow along on screen.
The text is taken out of the PDF on your device and spoken by a voice your browser already has. No account, no audio file to wait for.
What you get
- The sentence being read is highlighted, and in most browsers the word too. Click any sentence to read from there.
- Running headers, footers and page numbers are skipped, and words split across lines are joined, so it sounds like reading, not a list of lines.
- The language of the text is detected and a matching voice picked, including Hindi, Gujarati, Tamil and other Indian languages when your device has the voice.
- Speed from 0.5× to 2.5×, pitch and volume, changed while it reads.
- Pages to read such as 3-7, a stop timer, sentence and page skipping, and the time left.
- The cleaned text as a .txt file, and every page as a preview you can open full size.
What it can't do
- Save an MP3. Browsers speak the text but don't let a page record that voice.
- Read scanned pages. A scan is a picture of text; run PDF OCR first.
- Add voices. The voice list comes from your device and browser, so it differs between Windows, Android, iPhone and Mac.
- Keep the order of every layout. Tables, columns and side notes are read in the order the PDF stores them.
How to listen to a PDF in three steps
Add the text, pick a voice, press play.
Add a PDF or paste text
Drop a PDF or .txt file on the upload area, or paste text into the box below it and click Listen to this text.
Check the voice and speed
- A voice for the detected language is already chosen.
- Pick a speed chip, or Pages I type for part of the file.
- Your choices are remembered on this device.
Press play and follow along
Pause, skip a sentence or a page, or click any sentence to jump there. Space plays and pauses while the text has focus.
Settings that work for your situation
What to set before you press play.
| Situation | Why listening helps | What to set |
|---|---|---|
| Proofreading your own report | You hear missing words and repeated words your eyes skip. | Speed 0.75× or 1×, follow the highlight. |
| Revising study notes or a chapter | Hearing and reading together helps it stick. | Pages I type, for example 12-18; speed 1.25×. |
| Long article while cooking or walking | Your eyes are free. | Speed 1.5×, Stop after 30 minutes. |
| A Hindi or Gujarati circular | Easier for family members who prefer listening. | The language is detected; if no voice is found, follow the hint under Voice. |
| Dyslexia or tired eyes | The word highlight keeps your place. | Speed 0.75×, larger text with your browser zoom. |
| Bedtime reading | It stops by itself. | Stop after This page or 15 minutes, volume lower. |
| Checking a form or contract wording | Every line, page numbers included, can matter. | Untick Skip headers, footers and page numbers. |
The details
How the reading is worked out
What happens between your PDF and the voice.
Lines into sentences
pdf.js gives the text line by line. Lines are joined into paragraphs, a word broken with a hyphen at a line end is put back together, and your browser splits the paragraphs into sentences.
Headers and footers
A line at the top or bottom of the page that repeats on at least half the pages (3 or more) is treated as a header or footer. Lone numbers like 12 or Page 3 of 9 are skipped too.
Language and voice
The letters are counted by script: Gujarati, Devanagari, Tamil, Latin and others. The voice list is grouped by language, with the detected one first, and a matching voice is chosen.
Highlighting
Each sentence is spoken on its own, so the highlight always matches. Word marks come from the voice; some voices, like many online ones, only report whole sentences.
Pause and long text
Pause stops the voice and Play starts the same sentence again, which works the same in every browser. Very long sentences are spoken in parts, because some browsers stop a voice after about 15 seconds.
Where the text goes
The PDF stays on your device. Voices marked online in the list are run by your browser's maker, so the sentence is sent to their speech service; pick a voice without that mark to keep it on your device.
When the reading goes wrong
What usually causes it, and what to do.
"No text to read"
The pages are pictures of text, from a scanner or a phone camera.
Run PDF OCR, then paste the text here.
Play does nothing
The device is muted, or the browser blocked sound until you tap the page.
Turn the volume up, tap Play again. On iPhone, switch off silent mode.
Gujarati or Hindi read badly
Your device has no voice for that language, so an English voice tries.
Add the voice: Windows Settings, Time and language, Speech; Android Text-to-speech output; iPhone Accessibility, Spoken content. Or use Chrome or Edge.
Jumbled order
Columns, tables or text boxes are stored in a different order than you see.
Click the sentence where you want to continue, or paste the part you need.
Stops when the screen locks
Phones pause speech in background tabs to save battery.
Keep the screen on, then press Play; it carries on from the same sentence.
The PDF won't open
It needs a password to open, or it is damaged.
Use Unlock PDF or Repair PDF first.
Get a clearer reading every time
Small habits that make the voice easier to follow.
Try two or three voices
Voices differ a lot. Ones named Natural or Neural, or Google voices in Chrome, usually sound the most human.
Start at 1×, then speed up
Most people follow 1.25× to 1.5× easily after a few minutes.
Read only the pages you need
Choose Pages I type and enter a range like 4-9 to skip covers, contents and appendices.
Paste messy parts
For a table or a two-column page, copy the part you want and paste it into the text box.
Keep private files offline
For confidential documents pick a voice that isn't marked online.
Use headphones for proofreading
You'll catch small slips, like "the the" or a missing "not", more easily.
Common questions about the PDF Listener
If something's not covered here, the sections above likely have it.
Is my PDF uploaded to a server?
No. The PDF is opened and its text taken out in your browser, and it isn't uploaded to our server. The voice is your browser's own. Voices marked online in the list are run by your browser's maker, so each sentence is sent to their speech service; choose a voice without that mark to keep the text on your device.
Why can't it read my scanned PDF?
A scan is a picture of text, so there are no letters to read aloud. The tool tells you which pages have no text. Run the file through PDF OCR first, then open the copy here or paste its text.
Can it read Hindi, Gujarati or other Indian languages?
Yes, when your device has a voice for that language. The language is detected from the letters and a matching voice is picked. If none is installed, a note under Voice says how to add one: Windows Speech settings, Android Text-to-speech output, or iPhone Spoken content. Chrome and Edge also offer online voices for many languages.
Can I download the reading as an MP3?
No. Browsers speak the text with the device's voices but don't give web pages the audio to save. You can download the cleaned text as a .txt file instead.
Why is only the sentence highlighted and not each word?
Word highlighting needs the voice to report each word as it speaks. Most voices built into Windows, macOS, iPhone and Android do; some online voices only report whole sentences, so the sentence stays highlighted.
Why does it skip the header and page numbers?
Lines at the top or bottom that repeat on at least half the pages, and lone page numbers, are skipped so the reading flows. Untick Skip headers, footers and page numbers to hear every line.
What does it remember?
Your voice, speed, pitch, volume and the two tick boxes are saved in this browser on this device as soon as you change them. Your text and PDF are not saved; they are gone when you close the tab.
Build notes
How this PDF Listener works under the hood
I wrote up how to build a browser-based PDF reader that speaks: taking the text out with pdf.js and reading it with the Web Speech API, all on the user's own device.
How to Build a Browser-Based PDF Text-to-Speech Reader Using JavaScript
Covers text extraction with pdf.js, the speechSynthesis voice list, and playback controls in client-side JavaScript.