Skip to content

Privacy Policy & Disclaimer

Established: September 22, 2026 / Last updated: November 12, 2026

1. Data the app handles

Whisper Local (“the App”) processes the audio and video files the user chooses, and audio from the microphone the user chooses for live transcription, only on the user’s Mac. It does not send audio, recognition results, edits, history or the dictionary to servers of the operator (K386) or any third party. It has no feature for collecting usage statistics, crash reports or similar data.

Recognition results, edits, history and the dictionary are saved in ~/Library/Application Support/WhisperLocal on the user’s Mac and in the App’s settings. Temporary copies of audio requested by connected apps and activity records (logs) are also saved in the same folder. History can be deleted from “History” in the App’s settings.

In live transcription, the microphone is used only from when the user presses “Start” until the user presses “Stop”. By default, recordings (WAV) and transcribed text are saved in ~/Library/Application Support/WhisperLocal/Live. Whether to save them, where to save them, and automatic deletion (moving to the Trash) of items older than a set period can be chosen by the user under “Live” in the App’s settings. Even when recordings are set not to be saved, the most recent recording remains in a macOS temporary folder until the next recording or the next launch, so that “Transcribe Recording Accurately” can be used after stopping. Use of the microphone requires macOS permission, which can be revoked at any time in System Settings.

2. Network connections

The App communicates externally only in the following cases. In none of them does it send audio, recognition results, edits, history or the dictionary.

  • Model download: when the user instructs the App to download a model in the setup guide or in Settings, it downloads model files from the model distributor (Hugging Face). The distributor’s terms of use and privacy policy apply to this connection.
  • Update check: after launch and once a day, the App fetches update information (api/version.json) from this site. Only when a new version is available and the user chooses “Update and Restart” does it download the distribution file from this site. This check can be turned OFF with “Check for updates automatically” in Settings. Connections to this site are subject to the access logs described in “3. About this site” below.

Whisper speech inference runs in a separate process whose network access is blocked by an OS feature. When “Mac built-in” is chosen as the file transcription engine or the live transcription engine, the macOS speech recognition feature (provided by Apple) is used with on-device processing. If recognition data for that language is not on the Mac, macOS may download it from Apple. In that case, Apple’s privacy policy applies.

When the “Integration” feature is enabled, the App accepts connections only from inside the same Mac (127.0.0.1). It cannot be connected to from external computers. Credentials for connected apps are stored in the macOS keychain in a form that cannot be reversed (hash values). In single-confirmation connection (pairing) from a connected app, registration is performed only if “Connect” is chosen on the confirmation screen, and the credentials are not put in a URL but passed to the connected app over a connection within the same Mac. If “Launch at login” is turned ON, the App is registered in the macOS login items.

3. About this site

This site does not use cookies, advertising or analytics tools. It counts only page views and downloads, and does not store IP addresses (only the IP address ranges of connections blocked to prevent unauthorized access are recorded). For server operation, access logs (IP address, date and time, pages viewed, etc.) may be recorded for a certain period.

4. Disclaimer

The App is provided free of charge “AS IS”, without any warranty as to accuracy, completeness, usefulness, fitness for a particular purpose or safety. Speech recognition and speaker labels may contain errors. Check the recognition results yourself and use them at your own responsibility and judgment.

The operator is not responsible for any damages arising from the use of, or inability to use, the App or this site, including direct, indirect, incidental or consequential damages, loss or corruption of data, equipment failure, business interruption, loss of profits or any other damages.

For the audio you record or transcribe (including conversations recorded from the microphone in live transcription), you are responsible for complying with applicable laws, such as obtaining the consent of the people involved and handling copyright, portrait rights and personal information. Also check the results of automatic replacement by the dictionary and automatic insertion of punctuation yourself.

5. Changes

This policy may be changed without notice. The changed content takes effect when it is posted on this page.