We use a small number of trusted third-party services to deliver the product. Each is listed below with its purpose, what data it processes, and its hosting region. We update this page whenever a sub-processor is added, removed, or replaced.
| Provider | Purpose | Data processed | Region |
|---|---|---|---|
| Microsoft Azure App Service | Application hosting and request routing | HTTP requests, session cookies, server-side logs | South Central US |
| Microsoft Azure Storage (Tables) | Persistent storage for user records and monthly usage counters | User IDs, email addresses, plan + quota state | South Central US |
| Microsoft Azure AI Speech | Real-time speech-to-text transcription | Audio streamed live from the presenter's browser. Audio is not stored by us or — per Microsoft's terms — by Azure. | South Central US |
| Microsoft Azure AI Translator | Text translation between languages | Transcribed text segments only (no audio); processed in-memory and discarded after translation | South Central US |
| Microsoft Entra ID | Authentication for presenters (audience joiners do not authenticate) | OAuth identity (object ID), email address | Global Microsoft identity platform |
| Google (OAuth) | Optional "Continue with Google" authentication for presenters | OAuth identity (subject ID), email address | Global Google identity platform |
| Microsoft Azure Communication Services (Email) | Transactional email: sign-in links and quota notices | Recipient email address, message content (sign-in links, usage figures) | United States |
| Stripe | Subscription billing and payment processing | Email address, subscription state, payment card details (collected and stored solely by Stripe — card numbers never reach our servers) | United States |
Audio is never stored. The presenter's microphone stream is forwarded directly from the browser to Microsoft Azure AI Speech, transcribed in real time, and discarded. We do not extract or retain voiceprints. We do not record sessions.
Caption text is processed in memory for live broadcast to your listeners and is not retained after the session ends.