How to read a PDF aloud on Mac or Windows, free

The workflow: select inside your PDF viewer, then trigger

A lot of "read my PDF aloud" tools ask you to upload or import the file into their own app before anything happens. Shmia skips that step entirely, because it works on whatever text you've already selected, wherever it lives — including directly inside your PDF viewer.

On Mac, that's usually Preview, the app that opens PDFs by default. Open your PDF, click and drag to select a paragraph, a page, or several pages of text, then press Option+Command+R or click the small play bubble that appears next to your selection. On Windows, the same works inside Adobe Acrobat or Microsoft Edge's built-in PDF viewer — select text, press Ctrl+Alt+R or click the bubble. Either way, Shmia starts speaking in about a second, streaming the first sentences while it keeps generating the rest, so a 40-page report reads just as smoothly as a single paragraph.

A small floating player sits on top of your PDF viewer with a live timeline, play/pause, and a speed control from 0.75x to 2x — handy for slowing down a dense contract or speeding through a report you've mostly already skimmed. It never grabs focus away from the PDF window, so you can keep scrolling or highlighting while it plays.

The one real limitation: scanned pages aren't selectable text

It's worth being upfront about this rather than overselling: if a PDF is a scan — a photographed or scanned page saved as an image, common with older documents, some textbooks, and mailed contracts — there's no real text underneath for your viewer to select. You'll notice this immediately: trying to click-and-drag over the words either selects nothing or draws a rectangle around the image instead of highlighting individual words.

No selection-based tool, Shmia included, can read a page like that aloud, because there's no text to select in the first place — only pixels that happen to look like text. The fix is running the PDF through OCR (optical character recognition) first, which converts the scanned image into real, selectable text. Preview and Acrobat both have some OCR capability, and there are dedicated OCR tools for cases where the built-in options fall short. Once a PDF has been OCR'd, its text becomes selectable like any other document, and Shmia (or any other text tool) works on it normally.

Most modern PDFs — reports exported from Word or Google Docs, ebooks, most contracts, anything generated digitally rather than scanned — already have real selectable text and need no extra step at all.

Why this beats a dedicated "PDF reader" app

Plenty of tools exist purely to read PDFs aloud, but most want you to open the file inside their app specifically, which means leaving your normal PDF viewer, re-finding your place, and often losing your existing annotations or bookmarks view. Because Shmia works on a selection in whatever app already has the PDF open, there's no switching — you keep using Preview, Acrobat, or Edge exactly as you already do, and just add a shortcut press when you want to listen instead of read.

It's also free for this. Shmia's 10 local voices run entirely on-device and are unlimited, forever, with no per-page or per-minute cap — a genuine fit for reading a long PDF report, a research paper, or study material without watching a usage meter.