Privacy, in plain words.

This draft describes the current Spoken service for blog publishers and their listeners. The legal operator’s name, address, privacy contact, and effective date will be added before the service launches.

For publishers, Spoken processes account information such as your name, email address, profile image, sign-in provider, and session details. It also stores your site domain, player preferences, article URLs and titles, generated audio, and subscription information.

To generate audio and summaries, Spoken retrieves article content and sends the text to OpenAI. Audio files are stored using Vercel Blob. Do not submit confidential material or personal data you are not authorized to share.

If you contact us, we process your name, email address, optional website, and message so we can respond.

The player records playback milestones, listening mode, the article identifier, and event time. These listening events do not include an IP address, user identifier, or device identifier.

Requests use a daily-salted hash of the visitor’s IP address for rate limiting. This hash is kept separately from listening events. Rate-limit counters are periodically removed after their window expires.

Publisher sign-in uses session cookies. The marketing site stores your light or dark appearance preference in your browser. Hosting and other service providers may process technical request information in their own systems.

We use information to provide accounts, generate and deliver audio, manage subscriptions, display listening analytics, answer messages, and protect the service from misuse. Before launch, the operator will confirm the legal bases for these activities, including contractual necessity, legitimate interests, legal obligations, and consent where required.

The current implementation uses OpenAI for speech and summaries, Stripe for subscription payments, Vercel Blob for audio storage, and Resend for contact email when configured. Google or Microsoft handles sign-in when you choose that provider. Hosting and database providers will be identified before launch, together with relevant processing locations and international transfer safeguards. Spoken does not receive your complete payment card number from Stripe.

Account records, article metadata, audio, listening events, and contact correspondence may remain stored while they are needed to operate the service. The operator must set specific retention periods and a deletion process, including treatment of backups and required billing records, before launch. You can request access or deletion through our contact page once it is available.

Depending on the law that applies, you may have rights to access, correct, delete, restrict or object to processing, receive a portable copy of your data, and withdraw consent. You may also complain to your local data protection authority. Send requests through Contact; we may need to verify your identity.

Publishers are responsible for informing their own visitors about the embedded player and determining whether additional notices or consent are required for their site.

We will update this policy when our practices change and provide any notice required by applicable law. This draft has no effective date yet. Please review the final policy before using the launched service.