Last updated: 2026-11-10
Hear ("the App") is a real-time speech interpretation app. We do not operate servers that receive your audio or transcripts. You can use Google's Gemini API with your own key, or select a supported on-device translation experiment. Release builds use Firebase for crash reports and usage statistics. You can also choose to contact us by e-mail.
The App is published by 1XP Inc. ("we"). We do not operate any backend servers that receive your content, and we have no account system.
The following are stored only on your device (in the iOS Keychain or local app storage) and are never transmitted to us:
If you pair an Apple Watch, your API key and settings are synced directly to your Watch over Apple's WatchConnectivity (device-to-device). They do not pass through us.
The App does not create saved audio recordings or transcript/error diagnostic files, and has no diagnostic-recording or diagnostic-export feature. Audio and captions are processed in memory during interpretation. Web/media playback can use system caches; HLS playback also uses temporary segment files that are removed after decoding. These are not saved session recordings.
On upgrade, the App removes diagnostic files created by older versions from its own local storage. Cleanup is retried on launch/foreground and, on iOS, when protected files become available after unlocking.
App-written media logs contain media type and host metadata, not URL paths, query strings, credentials, audio, captions, or raw page error messages.
When a Gemini translation engine is selected, the App sends audio you capture (microphone, or media playing in the in-app browser/player) and the resulting text to the Google Gemini API, authenticated with your own API key. This is a direct connection between your device and Google under your own Google account/key.
This data is processed by Google under Google's Privacy Policy and the Gemini API Terms:
The App streams audio for interpretation and discards processing buffers. We never receive it. Google's processing and retention depend on your Google API account and the applicable Gemini API terms; the App does not control them.
The optional "AI Brief" feature sends the topic keywords you type to the same Gemini API (with Google Search grounding) using your key.
Supported Apple speech/translation and Apple Intelligence experiments process speech and text on the device. If you choose a Gemini speech renderer with a local translation engine, translated text is sent to Google to generate speech. Choosing a local translation engine does not make an explicitly selected remote speech renderer local. Optional model downloads contact Apple or the configured model host.
To find out why the App crashes or misbehaves, and which features are actually used, the App uses Google's Firebase Crashlytics (crash reporting) and Google Analytics for Firebase (usage statistics). This is everything that is sent to Google's servers:
For App Store disclosure, we treat the declared data as linked to an account or device: Gemini requests use your Google API credentials, and Firebase data can be associated through installation/device identifiers. This does not mean that we receive your audio or know your name. Google may de-identify content for particular uses; its account terms and processing rules still apply.
Never sent: audio, transcripts or translations, translation context, your Gemini API key, personal information such as name or e-mail, the advertising identifier (IDFA), precise location (there is no location permission). Nothing is used for advertising and nothing is linked across apps or websites. Firebase processing and retention are described in Privacy and Security in Firebase. Its Apple disclosure guide and Analytics disclosure guide describe SDK collection. Removing the App's recording feature does not disable Firebase crash reports or usage statistics.
"Send feedback" in Settings opens your own Mail app with the App version, device model and iOS version pre-filled in the body. Nothing is sent until you press send; when you do, that message (including your sender address and whatever you wrote) arrives at developer@1xp.com. We use it only to handle your request, do not keep it once the request is closed, and never pass it on. Write to the same address if you want it deleted.
The App uses the microphone only while you are actively translating, to capture speech for real-time translation. Microphone audio is not saved as a recording.
Hear is intended for adults aged 18 and over and is not directed to children. Gemini API use requires users to be at least 18. Store age classifications may vary by country or platform; an age classification does not waive Google's API requirements.
We may update this policy; the "Last updated" date will change accordingly.
Questions: developer@1xp.com