Dictation glossary

Push-to-talk

Push-to-talk means a recording or transmit channel that is active only while you hold a control, then stops when you release it. In dictation, that control is usually a keyboard key: hold to speak, release to finalize. Toggle mode is the alternative, where one tap starts listening and another tap stops. Both send audio to a recognizer; they differ in how you mark the start and end of an utterance.

Get WhisperJot Try the browser demo
In more detail

What is Push-to-talk?

What is push to talk as a UX pattern? Radios made it famous: you press a button on a handset, talk, and let go so others can reply. Dictation borrows the same muscle memory. Hold-to-talk is hard to leave on by accident, which limits capturing a private aside. It favors short bursts — a Slack message, a commit comment, a sentence in an email. The cost is that your finger stays busy for the whole utterance, which is awkward for a long paragraph.

Toggle, sometimes called tap-to-talk or hands-free start/stop, frees that finger. You tap once, speak at length, tap again. It is easier to ramble and easier to forget you are still recording. Some apps offer both on one key: a hold is push-to-talk, a short tap is toggle. Voice activity detection can trim leading and trailing silence in either mode, but it does not replace an explicit stop if you want a hard boundary before insertion.

For writers who speak

Why it matters for dictation

The wrong mode fights your writing rhythm. Short, frequent dictation wants push-to-talk so each thought is a bounded take. Long-form drafting wants toggle so you can pace, pause, and keep going. A global hotkey that supports both avoids training two different gestures. Safety matters too: a mode that cannot get stuck on is kinder in an open office.

In this product

How WhisperJot handles it

WhisperJot treats one trigger as both patterns: hold for push-to-talk, tap to toggle, including the fn key on Mac and a user-configured chord such as Option-Space or Control-Alt-Space. Live partials show in the HUD while you speak. Local engines stay the default; Jot Cloud is opt-in if you select it.

Related terms

Keep reading in the glossary.

Global hotkey

Global hotkey means a keyboard shortcut registered with the operating system so it works no matter which application is in front.

Read Global hotkey →

Voice dictation

Voice dictation is a way of writing in which you speak and software turns those words into editable text in an application.

Read Voice dictation →

Live partial transcription

Live partial transcription means showing in-progress text while audio is still being captured, instead of waiting until you stop talking to reveal a finished transcript.

Read Live partials →

Voice activity detection (VAD)

Voice activity detection (VAD) is a filter that decides whether a slice of audio contains speech or is silence, noise, or other sound.

Read VAD →
Questions

Straight answers.

See also: Dictation app for Mac.

What is push to talk?

Push to talk is a hold-to-record control. The microphone is live only while the key or button is down, and releasing it ends the utterance. In dictation, that end signal tells the app to finalize speech-to-text and insert the words. The pattern comes from two-way radios. It is the opposite of a toggle, where one press starts and a later press stops.

What is the difference between push-to-talk and toggle dictation?

Push-to-talk is hold to speak and release to stop. Toggle is tap to start and tap again to stop, so your hands can leave the key during a longer passage. Push-to-talk reduces accidental recording; toggle is more comfortable for paragraphs. Some dictation apps map both behaviors to the same physical key by distinguishing a hold from a tap.

Is push-to-talk more private than always-on listening?

Push-to-talk is a control pattern, not a full privacy design, but it does mean audio capture is gated on an explicit hold. That is stricter than an always-listening wake word. Privacy still depends on where the audio is transcribed — on-device or in the cloud — and on whether the app keeps a recording. A hold-to-talk cloud engine still sends audio while the key is down.

One hotkey, any focused app.

Private local transcription by default, with an optional opt-in cloud engine.

Get WhisperJot — $99/year