← Back to Sounds

Sounds for iMessage

Last updated: October 28, 2026

Overview

Sounds for iMessage (“the app”) is an iMessage extension that lets you browse and send audio clips in conversations. Your privacy matters. This policy explains what data we collect and how we handle it.

Data We Do Not Collect

We do not collect, store, or transmit any of the following:

The catalog does not require registration. Optional custom sounds use a pseudonymous library identity and a secret recovery key. Audio you choose to upload can contain personal information.

Data Stored on Your Device

The app stores a small amount of data locally on your device using standard iOS storage (UserDefaults). This includes:

Purchases

In-app purchases are processed entirely by Apple through the App Store. We use RevenueCat, a purchase management service, to verify purchase receipts and manage entitlements.

RevenueCat may receive:

RevenueCat does not receive your name, email, or any personally identifiable information from us. You can review RevenueCat’s privacy policy at revenuecat.com/privacy.

We occasionally run price experiments, so different users may be shown different prices for the same unlock. Which product (and therefore which price) you were shown is recorded in our anonymous analytics as a product ID so we can compare the variants. This is never linked to your identity.

Network Requests

The app connects to our server (hosted on Supabase) to fetch the latest sound catalog, trending sounds, and community sound requests.

Anonymous Product Analytics

The app also sends lightweight anonymous product analytics events to Supabase for app improvement, including:

These events may include an anonymous install ID, event timestamp, app version, optional sound/category IDs, the product ID of the unlock price you were shown (used to compare price variants), and a small numeric value where relevant (for example, the streak day count). They do not include names, email, phone number, contacts, message content, or ad identifiers.

Catalog Searches

Starting with version 1.1.0, settled searches in the public sound catalog may send the search term, result count, timestamp, and pseudonymous install ID to Supabase. We use these to find missing sounds and improve discovery. We do not send every keystroke or searches made on the personal-recordings tab. Terms are limited to 80 characters; obvious email addresses, URLs, and long numeric identifiers are excluded, but these filters cannot identify every kind of personal information. Avoid entering private information in catalog searches. Raw search rows expire after 90 days through daily maintenance. Search data is not public or used for advertising.

Sound Requests & Voting

If you use the sound-request feature, the sound or phrase you type is sent to Supabase along with the anonymous install ID. New request text is reviewed before it is shown to other users. Votes use the anonymous install ID only to prevent duplicate voting and apply rate limits. Do not include personal information in a sound request.

Custom Sounds

If you record or import a sound and choose to upload it, we collect the audio, a cover photo you select, its title, duration, file size, a pseudonymous creator/library ID, saved-library associations, share references, and reports or creator blocks. We store library metadata and private audio through Supabase; the existing public catalog uses Cloudflare R2. Vercel hosts the application API and sharing page. We may move private audio to a private R2 bucket without changing its sharing permissions.

Audio is validated and converted for playback. Selected cover photos are resized and stripped of embedded metadata. OpenAI receives uploaded audio for transcription and the title/transcript and selected cover photo for automated screening. Transcripts are not saved in our library database. Provider processing and retention are subject to the provider's policies: https://platform.openai.com/docs/guides/your-data. Automated screening is incomplete; report inappropriate sounds for staff review. Staff may access recordings for moderation or support, and moderation access is logged.

Creation requires an existing Unlock All Sounds purchase. We verify Apple-signed transaction proof and store the product, transaction identifier, environment, and revocation status linked to your pseudonymous library. Receiving, saving, and resending do not require a purchase. Your custom recordings do not appear in the public catalog automatically. Shared messages and links grant access to the recording; recipients can save, reuse, and forward it. Anyone with a shared link can listen through the sharing page. This hosting and processing is separate from Messages encryption. Do not use it for confidential recordings.

Recordings and drafts are also stored locally in the app's shared storage. Microphone access is requested only when you choose Record. We do not read conversation history, collect contacts, or record automatically. A recovery key grants access to your library, including deletion; keep it secret. It is stored in iOS Keychain and is copied only at your request. The app does not provide automatic account backup.

Custom-Sound Analytics

We record custom-sound saves, playback starts, and send starts with pseudonymous library/sound IDs and timestamps to understand reuse. These analytics do not contain raw audio, titles, transcripts, link secrets, or conversation content. Network providers may process IP addresses; the API uses hashed IP addresses for temporary abuse limits.

Deletion & Library Controls

You can remove a saved sound, delete recordings you created, block a creator, report a sound, and delete your custom library in Library Settings. Deletion stops future authorized access and triggers removal of hosted audio; already issued playback links may remain usable for up to five minutes. Copies previously exported or downloaded outside our app cannot be recalled. Abandoned uploads are eligible for cleanup after 24 hours. Minimal tombstones, abuse reports, and moderation records may be retained for security, deletion enforcement, and support. Contact us for data requests.

Children's Privacy

The app does not knowingly collect personal information from children. Sound requests are intended only for sound or phrase ideas and should not contain personal information.

Advertising & Tracking

No advertising SDKs, ad identifiers, or cross-app tracking frameworks are used.

Changes to This Policy

If we update this policy, we will revise the “Last updated” date above. Continued use of the app after changes constitutes acceptance.

Contact

If you have questions about this privacy policy, contact us at hello@jerryfeng.com.