Privacy

Our commitment to your privacy

MeetingDebrief is designed from the ground up to protect your privacy. We believe meeting recordings and transcripts are among the most sensitive data you can create, and we've built our architecture to ensure that data never leaves your control.

What we collect

Nothing. We do not collect or store your meeting data on our servers, and MeetingDebrief sends nothing elsewhere unless you connect something yourself. Specifically:

• No audio recordings are uploaded anywhere
• No transcripts or meeting summaries are sent to us
• No usage analytics or telemetry data is collected
• No crash reports are sent automatically
• No personal information is gathered

How it works

All processing in MeetingDebrief happens directly on your computer:

• Audio recording is captured and stored locally on your filesystem
• Transcription uses the Parakeet or Whisper speech recognition models running on your device
• Dictation is transcribed on your device and pasted into the app you are using; dictation audio and text are stored locally
• Summarization and chat use language models that run on your computer by default. If you connect your own AI server, they run there instead

The built-in AI models are downloaded once during initial setup and then operate completely offline.

Network access

By default, MeetingDebrief only makes network requests in these scenarios:

• Initial model download: When you first set up the app, AI models are downloaded from our CDN. After this, no network access is needed.
• License activation: When you enter your license key, the app verifies it with our server. This request only includes the license key and does not contain any personal information (only a unique identifier for your computer) or usage data.
• Update checks: The app periodically checks for software updates. No personal data is included in these requests.
• Feedback you choose to send: When you send feedback from the app, we receive what you write and any contact details or diagnostics bundle you choose to include.

Services you connect

Your audio is processed on your computer, always. Two optional features can send other data to a destination you choose. Both are off until you set them up:

• Your own AI server: transcripts, notes, and chat questions go to the server you configure. Audio does not. Access tokens are stored in your system's keychain.
• Integrations: webhooks receive details such as the title, time, length, folder, tags, and speaker names. Sending the summary or transcript is a separate choice for each integration. Scripts and Shortcuts run on your computer and receive the same event data. Webhook addresses and secrets are stored in your system's keychain.

Data sent to these destinations does not reach us. The operator of that server or service is responsible for how it handles the data, so pick one you trust.

Data storage

All your data (recordings, transcripts, summaries, dictations, and settings) is stored locally on your computer in the application's data directory. In Settings → Storage you can move recordings and Markdown files to a folder you choose, such as an external drive. You have full control over this data and can delete it at any time by removing the application, its data folder, and any folder you chose.

Third parties

We do not share any meeting data with third parties because we don't have any of it to share. There are no third-party analytics, advertising, or tracking services integrated into MeetingDebrief.

Changes to this policy

If we ever change this privacy policy, we will update the "last updated" date at the top and notify users through the application. Our commitment to local-first, privacy-respecting software will not change.

Contact

If you have questions about this privacy policy, please email us at [email protected].