On-device transcription
Use Apple’s on-device Speech engine, including the newer macOS 26 models, to transcribe locally so audio does not leave your Mac.
VTT is a native macOS dictation app for private on-device voice-to-text, with optional cloud engines using your own API keys.
VTT is a native macOS dictation app for people who want voice-to-text that stays close to the system experience. It supports on-device transcription by default, with optional cloud engines if you want to trade privacy for another accuracy path, and it is built only for Mac.
The app is designed around a menu-bar workflow: trigger dictation with a global hotkey, have text inserted into the app you are working in, and keep a local history of recent transcripts. The site also highlights per-language engine selection, downloadable models, and support for accents that built-in dictation may handle poorly.
VTT’s privacy model is straightforward. On-device dictation keeps audio on the Mac, while cloud engines are used only with your own provider key. The result is a dictation tool aimed at users who want a private default, a native macOS interface, and more control over which engine handles each language.
Use Apple’s on-device Speech engine, including the newer macOS 26 models, to transcribe locally so audio does not leave your Mac.
Run VTT from the menu bar with a global hotkey, a live waveform, and auto-insert into the app you are currently typing in.
Bring your own API key for Deepgram, OpenAI, or ElevenLabs, then choose the model you want from each provider.
Route different languages to different engines, either automatically or by manual selection, so each language can use the engine that fits best.
Pre-download supported language models so dictation starts as soon as you press the shortcut, even without waiting on a download.
Keep a local, newest-first transcript history on your Mac so you can paste a recent dictation again if you need to.
Dictate text into emails, documents, and other Mac apps using a global hotkey and automatic insertion, instead of switching to a separate transcription window.
Keep sensitive audio on your device by using Apple’s on-device Speech engine for drafts, notes, and other private dictation.
Switch to Deepgram, OpenAI, or ElevenLabs when built-in dictation struggles with a strong accent or a particular language.
Route languages separately when you work across more than one language, so each one can use the engine that handles it best.
Re-paste recent transcripts from local history if you paste into the wrong place, close the app, or want to reuse the latest dictation.
By default, VTT transcribes on-device using Apple’s Speech engine, so audio stays on your Mac. There is no account or analytics. If you enable a cloud engine, audio is sent directly to that provider using your own API key.
Yes. VTT is free to start, with no account required. On-device dictation costs nothing. Cloud engines are pay-as-you-go through your own provider API key if you choose to use them.
VTT supports Apple’s on-device Speech engine, including the macOS 26 models, plus optional cloud engines from Deepgram, OpenAI, and ElevenLabs. You can choose the engine per language.
VTT requires macOS 14 or later and runs on both Apple Silicon and Intel Macs. The site says it recommends the correct build automatically.
Yes. On-device dictation works fully offline. An internet connection is only needed for cloud engines or to download additional language models.
Traffic data is for reference only.
dictro.com
Dictro is a private AI dictation app for Mac that transcribes speech into polished text in any standard Mac text field. Keep the default workflow on-device, with optional cloud Fast Mode and local transcript history.
frigade.com
Yap is a free, open-source macOS dictation app that turns speech into text in any field you can type in, using on-device speech recognition and local transcript storage. No account, API key, telemetry, or network code.
spokenly.app
Spokenly is an AI dictation app for Mac, iPhone, and Windows that turns speech into text in any app. It supports free offline local models, cloud transcription through Pro or your own API keys, and more than 100 languages.
whisperstream.io
Whisperstream is a Windows dictation app that transcribes speech locally and inserts the resulting text into compatible apps. It is designed for people who want push-to-talk voice input without sending microphone audio to a cloud transcription service.
quickquill.app
QuickQuill is a macOS app for private, on-device dictation, transcription, and meeting notes. It offers free push-to-talk dictation plus a paid workflow for capture, file import, summaries, translation, and export.
wisproapp.com
Wispro is a native Windows voice dictation app that pastes transcribed or rewritten text into the active application. It helps you write faster with a hotkey, spoken input, and three workflow modes.