Clair Flow processes your audio in the cloud, so trust has to be explicit. This page describes what the product does with your data in plain language.
What Clair Flow stores
- account data such as your email address
- browser sessions, device approvals, and device records
- hashed device API keys
- subscription and billing identifiers
- usage counters and dictation session metadata
- transcript text associated with completed dictation sessions, kept until you delete it
What stays local to the Mac app
- microphone capture
- hotkeys and shortcut preferences
- local permission state
- text insertion into the focused app
What the cloud service processes
During dictation, Clair Flow sends audio to the cloud so it can:
- authenticate the connected device
- perform speech recognition
- clean and normalize the transcript
- apply glossary-aware terminology guidance
- record usage and account state
Transcript text and session metadata are stored. Raw audio is processed during your dictation session but is not saved afterward.
How long transcripts are kept
Clair Flow stores your transcript text so dictation sessions can be recovered if insertion fails or a connection drops mid-session. There is no automatic expiry window: your transcript text is kept until you delete it.
You can delete your dictation history at any time from your account dashboard, and that deletion removes the stored transcript text for those sessions. Raw audio is never stored, so there is no audio to delete — only the transcript text and session metadata. If you would like help clearing your history, email support@clairflow.ai.
Subprocessors
Clair Flow is a hosted product, and parts of the service are operated by third-party providers (“subprocessors”) that process data on Clair Flow’s behalf. Audio is sent to a speech-to-text provider during dictation and is not stored afterward; transcript text and account data are handled by the providers below for the purposes listed:
| Subprocessor | Role |
|---|---|
| Deepgram | Speech-to-text (real-time transcription) |
| AssemblyAI | Speech-to-text (real-time transcription) |
| Anthropic | LLM transcript cleanup |
| LLM transcript cleanup | |
| Cerebras | LLM transcript cleanup |
| Stripe | Subscription billing and payments |
| WorkOS | Account authentication and sign-in |
| Vercel | Account dashboard hosting and web session boundary (app.clairflow.ai) |
| Cloudflare | Hosting, CDN, and edge processing (Workers) |
| Convex | Application database (durable account and session state) |
Speech-to-text and transcript-cleanup providers may be used interchangeably depending on configuration and availability. Across all of them, raw audio is processed in transit during a dictation session and is not retained after the session ends.
Your data rights
Clair Flow is a paid product that processes voice input and the transcript text derived from it, so we keep your rights over that data clear. Depending on where you live, applicable laws such as the EU/UK GDPR and the California Consumer Privacy Act (CCPA/CPRA) may give you rights to:
- Access the personal data associated with your account.
- Delete your data — including deleting your dictation history directly from your account dashboard, or requesting account deletion by email.
- Export a copy of your data, available on request.
- Correct inaccurate account information.
To exercise any of these rights, use the controls in your account dashboard or email support@clairflow.ai. We do not sell your personal data, and we do not use your audio or transcript text to train third-party models beyond the processing needed to deliver the service described above.
What this means in practice
Clair Flow requires a cloud connection for speech processing and transcript cleanup. All dictation sessions are processed through the cloud pipeline described above.
Device security model
Clair Flow uses device-specific API keys for the native app:
- each connected device gets its own credential
- keys are shown once and stored hashed on the server
- keys can be revoked from the account dashboard
How the boundary works
The product is designed around a clean boundary:
- local app for capture and insertion
- cloud service for processing, billing, and account state
That boundary is central to how Clair Flow works today.