Text to speech reader: read PDFs and docs aloud free

Open the text where it already lives and use that app's reader. PDFs work well in Edge, documents in Word, articles in Chrome's reading mode, and on a phone you switch on one accessibility setting. All of them are free and none needs an account.

What they don't do is save the reading. It plays while the page is open, and once you close it, it's gone.

On a computer, the quickest route is Edge, which reads PDFs from its own toolbar:

1. 1Right-click the PDF and choose Open with > Microsoft Edge, or drag it into an Edge window.
2. 2Click Read aloud in the top toolbar, or press Ctrl + Shift + U.
3. 3Edge starts at the top and highlights each word; the toolbar buttons pause and skip paragraphs.
4. 4Use Voice options to change the speed or pick another voice.

The free ones all read on the spot. They differ in what they read, which devices they run on, and whether you can keep the audio.

If you'll hear something once while sitting at your computer, use the reader you already have. A news story, a long email or a quick check of your own paragraph doesn't need more.

An MP3 pays off when the text is long, when you'll listen more than once, and above all when you want to listen without a screen:

• Studying: a chapter turned into audio can be reviewed on the bus or on a walk, with your phone locked.
• Accessibility: the International Dyslexia Association lists text-to-speech among assistive technology tools for reading. A file plays in any audio app, with no need for someone to open the right page first.
• Proofreading: hearing your own draft in a natural voice exposes run-on sentences and repeated words your eyes skip over.
• Offline listening: an MP3 plays on a plane or the subway with no signal, and you can share it.
• Long sessions: a voice that sounds like a person is easier to listen to for an hour than the classic robotic one.

A reader speaks what's in the text, not what you see on the page. Converting a PDF to text pulls in things you never notice on paper but that break the flow out loud. Removing them takes five minutes:

• Headers and footers: the book or chapter title repeats on every page, and the reader will say it every time.
• Page numbers: they pop up in the middle of a sentence wherever the text crosses a page break.
• Footnotes and citations: delete them or move them to the end, or they'll interrupt the sentence they belong to.
• Columns: a two-column PDF can come out interleaved. Check that each paragraph still reads in order.
• Line-break hyphens: "know-ledge" split across lines is read as two fragments. Search for a hyphen followed by a line break and join them.
• Tables of contents, tables and long links: a TOC read aloud is two minutes of numbers, and a URL gets spelled out in full.
• Acronyms and abbreviations: they're read as written. If you want "e.g." or an acronym said a particular way, spell it the way it sounds.

New material: normal speed or slightly faster. Reviewing what you already know: up to 2×. In a 2022 study by Murphy and colleagues in Applied Cognitive Psychology, students watched lecture videos at 1×, 1.5×, 2× and 2.5×. Comprehension barely changed up to 2× and dropped beyond it. What helped most was using the saved time to listen again right before the test.

That study used lectures in the students' own language. If you're listening in a language you're still learning, start at 1× and work up.

Divide the character count by about 1,000 and you get a rough number of minutes at normal speed. Word shows it under Review > Word Count ("characters with spaces"); Google Docs under Tools > Word count. Besides length, the voice is what moves that number most.

To get the audio as a file:

1. 1Open the generator and click Upload document. It accepts .txt, .md, .srt, .pdf, .docx, up to 15 MB. The text is extracted in your browser; the file itself isn't uploaded to any server.
2. 2Check the text that appears in the box: remove headers, page numbers and footnotes as described above. You can also paste text directly.
3. 3Pick a voice and play its sample. For long listening, a calm narration voice works best. The language is detected automatically, and there are 7: English, Spanish, German, French, Portuguese, Italian and Russian.
4. 4Set the speed (0.5× to 2×) based on the table above.
5. 5Click Generate. You see the credit cost before anything is spent.
6. 6Play it in the browser and download the MP3 to your phone.

Sign in with Google and you get 600 free credits every month, about 10 minutes of audio: enough to try it on a chapter of your notes. They renew on their own and no card is needed.

After that, every 1,000 characters cost 60 credits, so a minute of audio is about 60 credits. There's no subscription, and purchased credits never expire.