Privacy Policy

Last updated: June 24, 2026

GlossOne is a native macOS app that transcribes speech and shows bilingual captions, using engines you choose yourself. GlossOne has no servers of its own, requires no account, and collects no personal data. This policy explains how your information is handled.

The short version

• Our App Store privacy label is Data Not Collected, matching the app's PrivacyInfo.xcprivacy manifest.
• We do not run any backend. GlossOne talks only to Apple and to the engines you set up.
• We do not collect, transmit, sell, or share your personal data.
• Speech recognition runs on-device, and translation can run on-device too. With those, your audio and text stay on your Mac entirely.
• If you pick a cloud translation engine, text goes directly from your Mac to that engine.
• Your API keys and secrets are stored in the macOS Keychain on your device.
• There is no analytics, advertising, or third-party tracking in the app.

Information we collect

None. GlossOne does not have an account system and does not send your data to us. We have no analytics SDKs, no crash-reporting services, and no advertising. The developer of GlossOne never receives your audio, your transcripts, your credentials, or usage information.

App Store privacy label: Data Not Collected

On the App Store, GlossOne's privacy label is Data Not Collected. The app collects no data from you and no data is linked to your identity. This matches the app's privacy manifest, PrivacyInfo.xcprivacy, which declares no tracking (NSPrivacyTracking is false), no tracking domains, and an empty list of collected data types (NSPrivacyCollectedDataTypes). The manifest also lists the required-reason system APIs GlossOne uses purely for its own on-device functionality, such as reading and writing its own settings and checking file timestamps for the recordings and models it manages on your Mac. Because GlossOne has no backend and no analytics, there is nothing for us to collect.

Audio, transcripts, and translations

GlossOne recognizes speech on your Mac on-device using WhisperKit, so the audio it listens to is processed locally and is not sent to us or to any third party. When you translate a transcript, GlossOne sends the relevant text directly from your Mac to the engine you configured (for example Apple's on-device translator, DeepL, Google, Microsoft, Tencent, Baidu, Youdao, Alibaba, Volcano Engine, Amazon, OpenAI, Claude, Gemini, or a local model via Ollama or LM Studio). There is no relay server in between. When you choose Apple's on-device translator or a local model, the text is processed on your Mac and is not sent to any third party. The handling of any data you send to a cloud engine is governed by that provider's own terms and privacy policy.

Credentials and secrets

API keys, access tokens, and secrets you enter for your engines are stored in the macOS Keychain on your device. They are used only to authenticate requests to the engine they belong to and are never sent anywhere else. They are not transmitted to the developer.

Data stored on your device

GlossOne keeps your settings, transcription history, and diagnostic logs locally on your Mac, inside the app's sandbox container. This data stays on your device. If you turn on optional audio recording, the recordings are saved to a folder you choose and stay there. You can clear your history and remove all stored keys or reset all settings from the app's preferences.

Microphone

Transcribing your own voice uses macOS Microphone permission. GlossOne requests it only when you start a microphone session, and you grant it in System Settings. The microphone audio is processed on your Mac to produce the transcript; it is not recorded, stored, or sent to the developer. The one exception is the optional audio recording you can turn on yourself, which saves to a folder you choose and stays there on your Mac.

Screen Recording (system audio)

Capturing the audio playing on your Mac, or in a specific app such as a meeting in Zoom, uses Apple's ScreenCaptureKit, which macOS gates behind the Screen Recording permission. GlossOne uses this permission only to read the audio from the source you pick; it does not record or transmit what is shown on your screen. As with the microphone, the captured audio is processed on your Mac to produce the transcript and is never sent to the developer. The permission is requested only when you choose a system or app audio source, and you grant it in System Settings.

Purchases

GlossOne offers a 7-day free trial and a one-time purchase to unlock the app, handled entirely through Apple's StoreKit and the App Store. Apple processes the transaction; GlossOne does not receive or store your payment details. Purchase validation uses Apple-signed receipts on your device. Please see Apple's Privacy Policy for how Apple handles App Store transactions.

Model downloads

The first time you use on-device recognition, GlossOne downloads the speech model it needs. That download comes from the model's hosting provider and contains no personal data about you. After it is cached on your Mac, recognition runs locally.

Network connections

GlossOne makes network connections only to (1) the translation engines you configure, in order to perform the actions you request, (2) the host that serves the on-device speech model, to download it once, and (3) Apple, for App Store and in-app purchase functionality. It makes no other outbound connections. If you point an engine at a local server (for example Ollama or LM Studio), that connection stays on your own machine or network.

Children's privacy

GlossOne is not directed to children and does not knowingly collect any information from anyone, including children under 13.

Changes to this policy

If this policy changes, the updated version will be posted on this page with a new “Last updated” date.

Contact

Questions about this policy or your privacy? Email support@glossone.one.