Speech to text on Mac has improved dramatically over the past few years, but the landscape of options is genuinely confusing. You have Apple's built-in dictation, a handful of third-party apps, browser-based solutions, and accessibility tools that overlap in their feature sets. Choosing the right approach matters because the wrong choice wastes your time and eventually causes you to abandon voice input altogether.
This guide covers everything you need to know about speech to text on Mac in 2026 — how the built-in option works, where it falls short, what to look for in a third-party app, and which workflows actually stick in daily practice.
Apple's Built-In Mac Dictation
Every Mac ships with a built-in dictation feature. You enable it in System Settings under Keyboard, then activate it by pressing the Function key or Globe key twice. A microphone indicator appears near your cursor and Apple's on-device speech recognition transcribes what you say.
What It Does Well
Built-in Mac dictation is free, private, and works offline once the language model is downloaded. For casual use — composing the occasional email or writing a quick note — it is completely adequate. The on-device accuracy for standard English prose is reasonable, and it supports basic punctuation commands like "period" and "new paragraph."
Where It Falls Short
The limitations become apparent quickly if you use dictation as a regular part of your workflow. The toggle activation model means dictation stays on until you press the key again, which leads to accidental transcription of ambient noise, other people's voices, and keyboard sounds. There is no hold-to-speak mode that stops listening the moment you stop pressing a key.
Accuracy with technical vocabulary, proper nouns, product names, and domain-specific terminology is noticeably worse than AI-powered alternatives. If you work in any specialized field — medicine, law, technology, finance — you will spend a significant portion of your time correcting misheard words.
The dictation experience also varies across applications. In native macOS apps with standard text fields, it works reliably. In Electron-based apps like Notion, Obsidian, VS Code, or Slack, activation can be inconsistent. In browser-based tools like Google Docs or Gmail, it works in some fields and not others.
What to Look for in a Third-Party Speech-to-Text App
When the built-in option does not meet your needs, you want a replacement that addresses its specific shortcomings. Here are the qualities that distinguish a genuinely useful speech-to-text app from one that merely sounds good in marketing copy.
Global Hotkey with Hold-to-Speak
The single most important feature is a hold-to-speak hotkey that works system-wide. You hold the key while speaking and it stops the moment you release. This prevents accidental transcription and creates a precise, controllable interface that feels like a push-to-talk radio. It should work in every application without exception — native apps, Electron apps, browsers, and the terminal.
High Accuracy on Specialized Vocabulary
A modern speech-to-text app should handle technical terms, proper nouns, brand names, and domain vocabulary without extensive customization. You should be able to dictate a software specification, a legal memo, or a clinical note and receive accurate transcription without babysitting every sentence.
Near-Instant Response
Latency kills the dictation habit. If you finish speaking and wait more than a second or two for the text to appear, the delay breaks your concentration. Look for apps that deliver transcribed text in under a second after you stop speaking.
Native Mac App, Not a Browser Extension
Browser extensions only work in your browser. System-level dictation requires a native macOS app that can inject text into any application's text field. A native app also integrates better with macOS security, uses less memory, and does not depend on a specific browser being open.
Steno: Speech to Text Built for Mac Power Users
Steno is a native macOS menu bar app that brings all of these qualities together. It uses a global hold-to-speak hotkey that works in every application on your Mac. You hold the hotkey, speak, release, and the text appears at your cursor — typically within half a second. No toggle, no mode switches, no waiting.
The accuracy on technical vocabulary is significantly better than Apple's on-device dictation. Steno handles software terminology, medical vocabulary, legal language, and proper nouns correctly in most cases without custom training. For specialized terms, you can add custom vocabulary in the settings.
Because Steno is a menu bar app, it is always available without taking up screen space. It works in Mail, Safari, Chrome, VS Code, Notion, Obsidian, Slack, Terminal, and every other text-accepting application on your Mac. There is no application-specific configuration required.
Practical Workflows for Mac Speech-to-Text
Email Composition
Open your email client, click in the body of a new message, and use speech to text for the body content. Keep the keyboard for the recipient and subject line. A well-structured email takes 20 to 30 seconds to dictate, compared to two to three minutes to type. Over the course of a day that includes 20 or 30 emails, this compounds into an hour of recovered time.
Document Drafting
For longer documents, use speech to text to generate a complete first draft before any editing. Speak continuously through each paragraph, releasing the hotkey between paragraphs to read back what you have said. This speaking-then-editing approach produces a first draft in roughly half the time of typing while also reducing the blank-page paralysis that makes starting difficult.
Slack and Chat
Short messages are actually one of the best use cases for speech to text because the ratio of words to message is high relative to the effort of typing. A Slack message that would take 30 seconds to type takes 8 seconds to speak. Multiply that by the dozens of Slack messages a busy knowledge worker sends each day.
Notes and Quick Capture
Keeping a notes app open in the background and using a global hotkey to dictate thoughts as they occur is one of the most powerful productivity habits enabled by system-level speech to text. The thought arrives, you hold the key, speak it into your notes app, and release. The capture takes less time than the thought itself.
Getting Started
If you are ready to try speech to text on Mac beyond what Apple's built-in feature offers, download Steno at stenofast.com. Setup takes about 30 seconds — install the app, grant microphone permission, set your hotkey, and you are ready to start dictating in any application. A free tier is available so you can test it with your actual workflow before subscribing.
The best speech-to-text tool for Mac is the one that gets out of your way. When dictation feels like dictating — not like managing a software tool — it becomes a habit that sticks.