Privacy Policy
Effective date: 25 July 2026
SpeakOS (“SpeakOS”, “we”, “us”) is a real-time AI speaking coach. This policy explains what information we collect, why we collect it, and the choices you have. It applies to the SpeakOS website and web app.
Information we collect
- Account information. When you sign up, our authentication provider (Clerk) collects your name, email address, and optional profile photo. If you sign in with a social account, we receive the basic profile details that provider shares.
- Diagnostic answers. Your responses to the onboarding speaking diagnostic (goals, experience level, focus areas), which we use to build your Communication Blueprint and recommend a learning path.
- Practice sessions. During a live coaching session your microphone audio is streamed to our voice infrastructure and transcribed in real time so the coach can respond. We store the session transcript and its duration to power your progress tracking. We do not store the raw audio recording after the session ends.
- Profile and settings. Details you choose to add, such as a bio, location, timezone, and notification preferences.
- Payment information. Payments are processed by PayChangu. We receive confirmation of the transaction (amount, currency, reference) but never see or store your card or mobile-money credentials.
How we use your information
- To run live coaching sessions and generate real-time feedback.
- To build your blueprint, track progress, and personalise your learning path.
- To process purchases and manage your plan and entitlements.
- To respond to support requests.
- To send service emails you have opted into, such as reminders and updates.
We do not sell your personal information.
How voice sessions work
Audio from your microphone is streamed over an encrypted connection to LiveKit (voice infrastructure), transcribed by Deepgram (speech-to-text), and the coach’s responses are generated by Google’s Gemini models and spoken back using text-to-speech from Deepgram or Cartesia. These providers process your audio and transcript solely to deliver the session.
Service providers
We share data only with the processors needed to run SpeakOS: Clerk (authentication and billing), LiveKit (real-time voice), Deepgram (speech-to-text and text-to-speech), Google (language models), Cartesia (text-to-speech), and PayChangu (payments). Each receives only what it needs to provide its service.
Data retention and deletion
We keep your account data, transcripts, and progress for as long as your account is active, because they power your growth tracking. You can request a copy of your data or the deletion of your account and all associated data at any time by emailing support@speakos.com. We will complete deletion requests within 30 days.
Security
Data is encrypted in transit, access to production systems is restricted, and authentication is handled by a dedicated provider. No system is perfectly secure, but we design for the safety of your data.
Children
SpeakOS is not directed at children under 16, and we do not knowingly collect their data. If you believe a child has created an account, contact us and we will delete it.
Changes to this policy
If we make material changes, we will update the effective date above and notify you in the app or by email before the changes take effect.
Contact
Questions about privacy? Email support@speakos.com.