# Text to Speech Mac: Every Built-In Way, Ranked > https://readaloudreader.com/blog/text-to-speech-mac/ — Turn on text to speech on a Mac with Option+Esc, download the Siri voices, read PDFs in Preview, and save audio with the say command. Published: 2026-09-09 · Author: Turan Zeynal · Category: Technology Every Mac already speaks. The feature has been buried in the accessibility settings since the OS 8 days, it has survived a dozen redesigns, and most people who own a MacBook have never switched it on. That is a shame, because once the shortcut is set up you can select any paragraph anywhere in macOS and hear it read back in under a second. Here is how the text to speech Mac ships with actually works, what its limits are, and when it makes sense to reach for a browser tool instead. If you mainly want to listen to documents rather than short selections, our guide to [turning a PDF into audio](/blog/turn-pdf-into-audio) covers that path in more detail. ## Switch on Spoken Content and pick a shortcut Everything starts in one panel. Open **System Settings → Accessibility → Spoken Content**. On older macOS versions the same panel lives in System Preferences under Accessibility → Spoken Content, and before that it was called Speech. - Turn on Speak selection. - Click the i button next to it to see the keyboard shortcut. The default is Option+Esc. - Optionally turn on Speak item under the pointer and Speak announcements if you want system alerts read too. Now highlight text anywhere — a web page, an email, a Numbers cell, a filename in Finder — and press Option+Esc. Press it again to stop. A small floating controller appears with play, skip and speed buttons, and you can drag it out of the way or collapse it to a single icon. That is the whole of it: Mac read aloud in its native form. No app to install, no account, works offline once the voice is downloaded. ## The default voice is not the good voice This is the part that trips people up. Out of the box your Mac speaks in a serviceable but flat system voice, and a lot of people try it once, decide macOS text to speech sounds robotic, and never come back. The better voices are a separate download. In the same Spoken Content panel, click the **System voice** dropdown and choose **Manage Voices**. You will find several tiers: - Siri voices — the same neural voices Siri uses, and by a wide margin the most natural option Apple ships. Each one is a few hundred megabytes. - Enhanced and Premium voices such as Ava, Allison, Tom and Samantha. Larger files, noticeably smoother than the compact versions of the same names. - Compact voices — the small default set, fine for short alerts, tiring over a long article. - Eloquence voices, added for screen reader users who prefer the fast, clipped classic style at high speeds. Download a Siri voice, set it as the system voice, and the difference is immediate. Speaking rate lives on the same screen; somewhere around 1.3x is where most people settle once their ear adjusts. ## Reading documents: Preview, Pages and Safari Selection reading covers most needs, but a few apps have their own speech menu that handles longer material better. In **Preview**, open a PDF and choose **Edit → Speech → Start Speaking**. It reads from wherever your cursor is, or reads your selection if you made one. The catch: Preview reads the text layer, so a scanned page produces silence. If that happens you need OCR first, which is exactly the problem our piece on [reading scanned PDFs aloud](/blog/read-scanned-pdf-aloud) deals with. The same **Edit → Speech** menu exists in TextEdit, Pages, Mail and most native apps, because it is provided by the system rather than the app. Safari has its own trick. Turn on Reader view with **Shift+Command+R** to strip navigation, ads and sidebars, then select all and hit Option+Esc. You get the article body without the voice announcing every menu item on the way down. ## Turning text into an audio file with the say command Worth knowing if you want something you can put on your phone. Open Terminal and run: - say "your text here" speaks immediately. - say -f notes.txt -o notes.aiff reads a text file and saves the result as an audio file instead of playing it. - say -v "?" lists every installed voice, so you can pick one with say -v Samantha. The output is AIFF by default. Add --data-format=alac and an .m4a extension if you want something smaller that syncs cleanly to an iPhone. It is the one corner of text to speech Mac power users actually love, useful for any macbook text to speech job, and it is free with the OS. ## Where the built-in system runs out of road Four limits show up quickly for anyone doing serious listening. **Voices are per-machine.** Every text to speech Mac setting you change lives on that Mac. Switch to a work laptop or a phone and you start again with the compact defaults. **No sentence highlighting.** The floating controller shows no text. Following along visually while listening, which is the single biggest comprehension aid for dyslexic and ADHD readers, simply is not part of the feature. **Long documents are awkward.** Speech stops at the end of a selection, so a 40-page report means selecting section by section. There is no chapter list, no resume-where-you-left-off, no queue. **Storage.** Two or three premium voices in different languages will happily eat a couple of gigabytes on a 256GB machine. ## When a browser tool is the better answer None of the above matters if you only ever read a paragraph at a time. It matters a lot if you are getting through papers, reports or long articles every day. That is the gap [Read Aloud Reader](/) fills: paste text or drop in a PDF, DOCX or EPUB, pick a voice, and listen with the current sentence highlighted as it plays. It runs in any browser, so the same setup follows you from a MacBook to a work PC to a phone, and you can export the audio as an MP3 to finish a chapter on a walk. Our comparison of the [most realistic text to speech voices](/blog/most-realistic-text-to-speech) is a reasonable place to judge whether the voice quality clears your bar before you try it. The honest recommendation is to use both. Option+Esc with a Siri voice is unbeatable for a quick "what does this sentence actually say" check, and it costs nothing to set up. For anything longer than a screen, a tool built around reading queues and highlighting will save you far more time than it takes to learn. Read Aloud Reader handles that half of the job, and neither one stops the other from working. ## Frequently asked questions ### How do I turn on text to speech on a Mac? Open System Settings, then Accessibility, then Spoken Content, and switch on Speak selection. Highlight any text and press Option+Esc to hear it. Press the same shortcut again to stop. ### Why does my Mac's voice sound robotic? You are hearing a compact system voice. Open Spoken Content, click the System voice dropdown, choose Manage Voices, and download a Siri or Premium voice. They are several hundred megabytes each but sound dramatically more natural. ### Can Preview read a PDF aloud on Mac? Yes. Open the PDF in Preview and choose Edit, then Speech, then Start Speaking. It only works on PDFs with a real text layer, so scanned pages produce no sound until they have been run through OCR. ### How do I save spoken text as an audio file on a Mac? Use the say command in Terminal: say -f notes.txt -o notes.aiff reads a text file and writes the audio to disk instead of playing it. Add --data-format=alac with an .m4a extension for a smaller file. Try the tool: https://readaloudreader.com/studio