Privacy — especially your child's voice
Children's voice recordings are among the most sensitive data a product can hold. SayBright is designed around that fact, not despite it.
Consent first, always
No recording is ever made before a parent or legal guardian accepts the consent step in the app. Consent is versioned and recorded; withdrawing it stops recording immediately. We design for COPPA (US) and equivalent child-privacy rules in other markets.
Who can hear a recording
Only your family account and a clinician your family has linked to your child. Only a parent can create or revoke that link. Clinicians in other workspaces can never see, hear, or even learn that your child exists — workspace isolation is enforced in the database, not just the interface.
Retention and deletion
Recordings are automatically deleted after 90 days; scores and progress history remain so your child's record is continuous. At any time, a parent can delete their child and all of their data — recordings, attempts, sessions, links and consents — in one action from the parent dashboard. Ending a trial or subscription never deletes anything.
What we never do
We never sell child data. We never use recordings for advertising. We never train shared models on your child's voice without separate, explicit opt-in consent (and the default is no).
Scoring processors
When full per-sound scoring is connected for your account, short audio clips are sent to our speech assessment provider for scoring and are governed by that provider's enterprise data terms. The built-in practice engine processes audio features on the device and sends nothing anywhere.
Questions: hello@saybrightapp.com. This page is the plain-language summary of our practices; the full policy ships with account signup.