Legal

Privacy Policy

Last updated: June 15, 2026

The short version

Your audio recordings never leave your device.

Transcript text is stored locally. Once per week, it's sent to generate your AI summary — then deleted from servers with zero retention.

No account required. We don't know who you are.

Core features work in airplane mode. Verify it yourself.

No advertising. No analytics SDKs that read your content.

1. Who We Are

Spoke is an AI-powered voice journal app developed independently. We're committed to building tools that respect your privacy as a first principle — not as a compliance checkbox.

If you have questions about this policy, contact us at shivamkeshri7044@gmail.com.

2. Your Audio

Audio recordings never leave your device. Full stop.

When you record a journal entry, Spoke uses your device's built-in speech recognition to transcribe your voice on-device. The audio file is stored locally on your phone. We do not upload, stream, or transmit audio to any server — ours or anyone else's.

Proof: Try recording in airplane mode. Core journaling features work fully offline because no audio data is sent anywhere.

3. Transcripts and the Weekly AI Summary

Your journal transcripts (the text version of what you said) are stored in a local SQLite database on your device.

Once per week, Spoke sends the text of your recent journal entries to generate your AI mirror summary. Here is exactly what happens:

Your transcript text (no audio, no name, no account identifier) is sent to a Cloudflare Worker relay server.
The relay forwards the text to Anthropic's Claude API with zero-data-retention enabled — meaning Anthropic does not store or use your content for training.
The AI summary is returned, stored locally on your device, and the relay discards the data.

We do not log, store, or associate your transcript content with any identifier on our servers. The relay exists solely to protect your API key, not to collect your data.

4. What We Do Not Collect

No name, email, or account information — no account is required to use Spoke.
No advertising identifiers or behavioral tracking.
No analytics SDKs that read the content of your journal entries.
No persistent device identifiers linked to your identity.
No audio recordings, ever.

5. Subscriptions and Payments

Spoke uses RevenueCat to manage subscription billing. Payments are processed by Google Play and are subject to Google's privacy policy. RevenueCat receives your subscription status (active/expired) and app-level identifiers for purchase verification. RevenueCat does not receive your journal content.

We never see or store your payment card information. All billing is handled by Google Play.

6. Local Data Storage

All journal entries, mood logs, streak data, and summaries are stored in a SQLite database on your device. This data is backed up as part of your standard Android backup (to your Google account) if you have that enabled. You can disable this in your Android settings.

Deleting the app removes all local data permanently.

7. Notifications

Spoke can send you a daily journaling reminder and a weekly summary notification. These are local notifications generated on-device and are not delivered through an external push notification service that tracks delivery or engagement. You can disable or reschedule them at any time in the app's Settings screen.

8. Children's Privacy

Spoke is not directed at children under 13. We do not knowingly collect information from children. If you believe a child has used Spoke, please contact us and we will take appropriate steps.

9. Changes to This Policy

We will notify you of material changes to this privacy policy through the app. The "Last updated" date at the top of this page reflects when the policy was last revised. Continued use of Spoke after changes constitutes acceptance of the updated policy.

10. Contact

Questions, concerns, or requests about your data are welcome.

Email: shivamkeshri7044@gmail.com