Last updated: October 7, 2026

Privacy Policy

This policy describes how Robin Food (“we,” “us”) handles information when you use the Speaking Reflex Chrome extension and this website. Speaking Reflex is in private beta; features and providers may change as the product evolves.

What Speaking Reflex does

Speaking Reflex helps you practice spoken English while you watch YouTube. You listen to a short segment, retell the idea in your own words, receive meaning-focused feedback, and may continue with grounded follow-up questions. The extension runs on YouTube pages you visit; AI features require signing in.

Information we collect

Depending on how you use the product, we may process:

• Account information — such as your email address, account identifier, and name if Google provides it through sign-in (via Supabase Auth).
• Authentication data — session tokens stored in the extension so you can call our API while signed in.
• Usage and billing metadata — plan status, AI practice session counts, session identifiers, subscription and customer references from our payment provider, and related records in our database. We do not store a history of what you said in practice in our database.
• YouTube captions (on your device) — caption text and related metadata cached locally in your browser for videos you practice with.
• Technical data — standard request metadata in our API logs (for example timing and diagnostic fields). We do not intentionally log full practice transcripts in these logs.
• Beta waitlist (website) — if you submit your email on our landing page to request notification when private beta access becomes available, we store that address to contact you for that purpose. This is not a general marketing newsletter.

Audio and speech processing

When you record a retell or follow-up answer, the browser captures microphone audio on the YouTube page. Audio is sent to our API for speech-to-text processing, then processed for evaluation. Our servers do not store your raw audio recordings after the request completes. Transcribed text and feedback appear in the extension UI during your session and are not saved to our database as practice history.

YouTube content and context

Caption tracks are fetched and cached on your device (IndexedDB) so practice can be grounded in what you heard. We do not send full video libraries or wholesale transcript dumps to our backend. For AI features we send only the relevant caption text window (and timing metadata needed for practice), not YouTube video IDs or URLs. Your use of YouTube remains subject to Google’s and YouTube’s terms and policies.

Account and authentication

Sign-in uses Google OAuth through Supabase. The extension stores your Supabase session in Chrome local storage so you stay signed in. You can sign out from the extension, which clears the local session. There is no in-extension button to delete your entire account today.

AI processing

We use OpenAI models configured by Speaking Reflex for speech-to-text, semantic evaluation, and follow-up question generation. Those services receive the audio or text needed to perform the feature you requested. We do not publish specific model names in this policy because they may change as we tune the product.

Billing and subscriptions

Pro subscriptions are sold through Polar, which acts as merchant of record for checkout and subscription management. When you upgrade, your email (and, at checkout, your IP address as provided by our API) may be sent to Polar. We store subscription status and provider identifiers in our database to enforce your plan and usage limits. Payment card details are entered on Polar’s hosted checkout, not in Speaking Reflex.

If you are signed in to the extension and ask to be notified when Pro becomes available, we use your account email (or the address you provide in that flow) to send that Pro availability notification. This is separate from the public beta waitlist and from a general marketing newsletter.

How we use information

We use the information above to:

• Provide speaking practice, feedback, and follow-up features you request.
• Authenticate you and keep your session secure.
• Enforce free and Pro usage limits and technical abuse safeguards.
• Process subscriptions and show your plan status in the extension.
• Operate, debug, and improve the service (including aggregated diagnostics).
• Respond to support and legal requests.

Service providers

We rely on trusted providers to run Speaking Reflex, including:

• Google — OAuth and YouTube (content you choose to watch).
• Supabase — authentication and hosted Postgres database.
• OpenAI — speech-to-text and language-model features described above.
• Polar — checkout, subscriptions, and customer portal.
• Cloudflare — API hosting, database connectivity, and platform logs.

Each provider processes data under its own terms and privacy practices. We use them only to deliver and support Speaking Reflex.

Data retention

We keep information only as long as needed for the purposes described in this policy:

• Practice audio and learner transcript text on our servers — not stored after the API request finishes; session UI data clears when you close practice or navigate away.
• YouTube caption cache on your device — remains until you clear site data for YouTube, change cache versioning, or remove the extension.
• Account, usage, and billing records in our database — kept while your account is active and as needed for billing, fraud prevention, and legal obligations. We do not currently run automated purges of historical usage or session rows.
• Checkout intents — may expire based on stored expiration timestamps and consumption flags.
• Provider and platform logs — retention is governed by Supabase, OpenAI, Cloudflare, Polar, and Google settings applicable to our accounts, not by a separate schedule published here.

Security

We use HTTPS for API traffic and verify authenticated requests with industry-standard tokens. No system is perfectly secure; please use a strong Google account and keep your browser up to date.

Sharing, sale, and advertising

We do not sell your personal information. We do not use Speaking Reflex data for unrelated advertising. We share information with service providers as described above, when you direct us to (for example opening Polar checkout), or when required by law.

Your choices and requests

You can:

• Sign out of the extension to clear the local auth session.
• Clear YouTube site data to remove locally cached captions.
• Cancel Pro through the Polar customer portal when available in the extension.

To request access, correction, or deletion of personal information we control, email robinfood.dev@gmail.com from the address tied to your account. We handle deletion manually (there is no automated self-service deletion flow). We aim to complete verified deletion requests within 30 days, subject to legal or billing records we must retain.

Children

Speaking Reflex is not directed at anyone under 18. You must be at least 18 years old to use the service. If you believe a child has provided us information, contact us and we will take appropriate steps.

Changes

We may update this policy from time to time. We will post the revised version on this page with a new “Last updated” date. Continued use after changes means you accept the updated policy.

Contact

Operator: Robin Food
Email: robinfood.dev@gmail.com

Chrome Web Store limited use

The Speaking Reflex extension uses Google user data and Chrome permissions only to provide speaking practice, account sign-in, and subscription flows described in this policy—not for unrelated advertising, creditworthiness, or sale of user data. Permissions such as access to YouTube, identity, storage, tabs, and text-to-speech exist to inject practice UI, maintain your session, open checkout, and read follow-up questions aloud locally.