Glossary
Dictation, plainly defined.
The vocabulary of speech-to-text, one page per term: what it means, why it matters in practice, where the marketing tends to blur it, and where you can see it working.
On your Mac
The settings, keys and features in macOS that dictation runs into, and where each one is.
Globe keyFn key, Function key
The Globe key is the key marked with a globe symbol, and usually also with fn, on Mac keyboards.
Read the definition
Voice ControlMac voice commands
Voice Control is an accessibility feature built into macOS that lets you operate a Mac entirely by voice.
Read the definition
Dictation commandsMac dictation commands, voice punctuation commands
Dictation commands are the words and phrases you say while using Dictation on a Mac to enter punctuation, symbols, capital letters, and formatting instead of ordinary text.
Read the definition
Auto-punctuationautomatic punctuation
Auto-punctuation is a setting of Dictation on Mac that automatically inserts commas, periods, and question marks as you dictate.
Read the definition
Input volumemicrophone volume, input level, mic gain
Input volume is the setting on a Mac that controls how strongly sound from the selected microphone or other input device is captured.
Read the definition
Mic ModeMic Modes, Voice Isolation, Wide Spectrum
Mic Mode is a macOS control that changes how your Mac’s microphone treats the sound around you in apps that support it.
Read the definition
Secure Keyboard EntrySecure Event Input, secure input
Secure Keyboard Entry is a Terminal setting on a Mac that prevents other apps on your computer or the network from detecting and recording what you type in Terminal.
Read the definition
Live Captionsreal-time captions, live transcription
Live Captions is a macOS accessibility feature that shows a real-time transcription of spoken audio in a window on your screen, for audio in any app and for live conversations around you.
Read the definition
Enhanced DictationUse Enhanced Dictation, offline dictation
Enhanced Dictation was a checkbox in the Dictation pane of Keyboard preferences on older versions of macOS that downloaded speech recognition software to the Mac so you could dictate offline.
Read the definition
Improve Siri & DictationShare Audio Recordings, Siri & Dictation History
Improve Siri & Dictation is an opt-in privacy setting that lets Apple store and review a sample of the audio from your Siri and Dictation interactions to develop and improve those features.
Read the definition
Apple Intelligence
Apple Intelligence is Apple’s set of generative AI features built into macOS, including Writing Tools for rewriting, proofreading and summarizing text.
Read the definition
SpeechAnalyzerSpeechTranscriber, Speech framework
SpeechAnalyzer is a class in Apple’s Speech framework, introduced with macOS 26 and iOS 26, that apps use to turn spoken audio into text.
Read the definition
Talking to it
The words for how dictation starts, stops, and reaches the app you are in.
Dictation
Dictation is producing written text by speaking, with the words transcribed in or near real time at the place you are writing.
Read the definition
Voice typing
Voice typing is another name for dictation: entering text into an application by speaking instead of using the keyboard.
Read the definition
Transcriptionfile transcription
Transcription is converting an existing audio or video recording into text after the fact.
Read the definition
Push-to-talkhold-to-talk
Push-to-talk is an activation style where recording runs only while a key is held down: press, speak, release, done.
Read the definition
Voice activity detectionVAD
Voice activity detection (VAD) is the technique of detecting when someone is speaking versus when there is silence or background noise.
Read the definition
Wake wordhotword
A wake word is a spoken phrase that turns a listening device on, like "Hey Siri" or "Alexa".
Read the definition
Accessibility permissionAccessibility access
Accessibility permission is the macOS setting that lets one app read and control the interface of other apps.
Read the definition
How it hears you
The models, the measurements, and the failure modes of turning audio into words.
Speech-to-textSTT
Speech-to-text (STT) is the technology that converts spoken audio into written words.
Read the definition
Automatic speech recognitionASR
Automatic speech recognition (ASR) is the machine-learning field, and the class of models, that turns audio of human speech into text.
Read the definition
Whisper
Whisper is a family of open-source speech recognition models released by OpenAI in 2022, trained on 680,000 hours of audio and covering about 99 languages.
Read the definition
Parakeet
Parakeet is a family of open-source speech recognition models from NVIDIA, built on a transducer architecture, which can emit words as audio arrives.
Read the definition
Neural EngineANE
The Neural Engine is the part of an Apple silicon chip built to run machine-learning models efficiently, alongside the CPU and GPU.
Read the definition
Streaming transcriptionreal-time transcription
Streaming transcription shows words on screen while you are still speaking, rather than after you stop.
Read the definition
Language identificationLID
Language identification is the step where speech software works out which language is being spoken before, or while, it transcribes.
Read the definition
Word error rateWER
Word error rate (WER) is the standard measure of speech recognition accuracy: the number of words a model got wrong, divided by the number of words actually spoken.
Read the definition
Hallucination
In speech recognition, a hallucination is text the model produces that nobody said: a plausible phrase invented from silence, noise, or a fragment too short to transcribe.
Read the definition
Latencykey-up latency
In dictation, latency is the time between the end of your speech and finished text appearing at the cursor.
Read the definition
Speaker diarizationspeaker labels
Speaker diarization is the process of working out who spoke when in a recording, so a transcript can be labeled by speaker.
Read the definition
Whisper large-v3 turboWhisper turbo, large-v3-turbo
Whisper large-v3 turbo is a speech recognition model from OpenAI, released on 1 October 2024.
Read the definition
Cohere Transcribecohere-transcribe-03-2026
Cohere Transcribe is an open-source speech recognition model from Cohere, released in March 2026.
Read the definition
Qwen3-ASRQwen3-ASR-1.7B, Qwen3-ASR-0.6B
Qwen3-ASR is a family of open-source speech recognition models from Alibaba’s Qwen team, released on 29 January 2026.
Read the definition
MLXApple MLX, MLX framework
MLX is an open-source array framework for machine learning on Apple silicon, made by Apple’s machine learning research team.
Read the definition
Core MLCoreML, Apple Core ML
Core ML is Apple’s framework for running machine learning models inside an app, on the device.
Read the definition
Quantizationmodel quantization, 4-bit, 8-bit
Quantization is storing a model’s numbers with fewer bits each, for example 4 or 8 bits where the original used 16 or 32.
Read the definition
Real-time factorRTF, RTFx
Real-time factor (RTF) is the time a speech recognition system takes to process audio, divided by the length of that audio.
Read the definition
What comes out
Everything that happens between a raw transcript and text you would have typed.
Disfluencyfillers
A disfluency is any break in fluent speech: fillers like "um" and "uh", repeated words, false starts, and self-corrections.
Read the definition
Formatting passpost-processing
The formatting pass is the step after speech recognition that turns a raw transcript into text a person would have typed: punctuation, capitalization, numbers as digits, fillers removed, and a tone suited to where the text is going.
Read the definition
Verbatim mode
Verbatim mode is a dictation setting where the output is exactly what you said, with no rewording, tone adjustment, or cleanup beyond basic recognition.
Read the definition
Spoken punctuationvoice commands
Spoken punctuation is saying the name of a mark or symbol, like "comma", "open paren", or "new paragraph", and having the dictation tool type the symbol instead of the word.
Read the definition
Text expansionsnippets
Text expansion replaces a short trigger with a longer saved text: an address, a signature, a canned reply.
Read the definition
Personal dictionarycustom vocabulary
A personal dictionary is the list of words a dictation tool has been taught to spell your way: names, products, jargon, and anything the speech model would otherwise write as the most probable English.
Read the definition
Tone matchingregister
Tone matching is a formatting feature that writes the same dictated words differently depending on where they are going: lowercase and loose in a chat app, punctuated and complete in an email, untouched in a terminal.
Read the definition
Words per minuteWPM
Words per minute (WPM) measures text production speed.
Read the definition
Where it runs
The two words that decide whether your voice becomes someone else’s data.
On-device processinglocal processing
On-device processing means the speech recognition model runs on your own computer, so audio is transcribed locally and never uploaded.
Read the definition
Local LLMon-device language model
A local LLM is a language model that runs on your own computer rather than on a vendor’s server.
Read the definition
Your voice was always faster.
7 days free, then $19 USD once. No account, no subscription, nothing uploaded.
macOS 14 or later · Apple silicon