Privacy
Your voice is processed on your device. Never uploaded, never stored, never used for training.
WhisperVoca is built so that privacy is the default, not an upgrade. Here is exactly what happens, and what does not happen, with your voice and your words.
Your voice never leaves your device
All transcription happens locally using Whisper AI. Audio is not uploaded to any server. There is no cloud step in dictation.
No audio recordings are stored
Your speech is transcribed and the audio is discarded immediately. The only thing kept is the text history you choose to keep, locally on your machine.
We never train AI on your voice
Your voice and what you say are never used to train or improve our models. Local models run on your computer. They do not learn from your speech.
Accounts are for login only
If you create an account, it is only for signing in and managing access. It is not used to collect your voice, store recordings, or train AI.
How it works in practice
When you hold your hotkey and speak, WhisperVoca captures audio from your microphone and feeds it to a Whisper speech model running on your own computer. The model returns text, the text is pasted into your active app, and the audio is gone.
Because there is no cloud component for transcription, there is no server that could log, store, or train on your voice. Speech models are downloaded once and then run locally, which is also why dictation keeps working with no internet connection.
Many cloud dictation apps upload audio to remote servers. Some keep recordings or use speech data to improve their models. WhisperVoca does none of that. Your voice stays on your device.
Frequently asked questions
Does WhisperVoca send my voice to the cloud?
No. WhisperVoca transcribes your voice entirely on your device using local Whisper models. No audio is uploaded to any server.
Are my recordings stored anywhere?
No. Audio is processed and discarded immediately after transcription. Only the text history you choose to keep is stored, and it stays on your device.
Do you use my voice to train AI?
No. WhisperVoca never sends your voice to our servers, so we have nothing to train on. Models run locally on your device and are not updated from your speech.
Why might I need an account?
Only to sign in and manage access if you choose to create one. Dictation does not need an account. Your voice is transcribed on your device either way, and is never uploaded or used to train AI.
How is this different from cloud dictation?
Many cloud dictation apps upload audio to remote servers. Some keep recordings or use speech data to improve their models. WhisperVoca does none of that. Transcription is local, audio is discarded, and your voice is never used for training.