The short version. Vocare Translate turns speech into translated speech. To do that, the audio you record while holding the speak button is sent to our server and to our speech and translation providers, converted to text, translated, and spoken back. There are no accounts, no sign-in, and no analytics or tracking SDKs. The Android app is free and shows a short advertisement from Google AdMob when you open it; section 2 explains exactly what that involves and how to limit it. Conversation transcripts are kept on your own device. A copy of the transcript text also exists on the server for the life of the running session process. Read on for exactly what that means.
1. Who we are
This app is published by Vocare AI Pty Ltd (ACN 696 851 635, ABN 63 696 851 635), referred to below as "we" or "us". We are the data controller for the personal information described in this policy.
2. What we collect and why
Microphone audio
When you hold the speak button, the app captures audio from your device microphone and streams it to our server over an encrypted WebRTC connection. This is the core function of the app: we cannot translate speech we cannot hear. The microphone is not opened when you are not in a translation session, and the app does not record in the background. On the Apple Watch companion, audio is recorded only for the duration of a single press and is uploaded as one complete recording when you release.
Transcripts of what is said
Your speech is converted to text (the "original") and that text is translated into the other language (the "translation"). Both texts are shown on screen, spoken aloud, and stored as the session transcript.
Language and session settings
The two languages you have selected, an optional display name for each speaker, an optional conversation topic, a randomly generated session code, and the session start time and duration. The display name and topic are free-text fields you fill in yourself — please do not enter information you would not want in a transcript.
Technical connection data
Establishing a real-time audio connection requires your device's network (IP) addresses to be exchanged with our server and with the relay service described in section 3. Our server writes operational logs — timestamps, session codes, connection events, recording sizes and audio levels, and error messages. These logs are for keeping the service working and are not used to profile you.
Advertising (Android app only)
The Android app is free and is supported by advertising. When you open the app, or bring it back to the foreground after a while, Google AdMob may show a full-screen advertisement that you can close using the control Google draws on it. We never show an advertisement while a translation session is in progress, and subscribers will not see advertisements once a paid tier exists. To serve and measure that advertisement, Google receives your device's advertising ID, your IP address (from which a coarse location is derived), basic device and app information, and whether you viewed or tapped the ad. Google never receives your microphone audio, your transcripts, or your translations. Where consent is required (the EEA, the UK and Switzerland) a consent form is shown before any ad is requested, and you can change your choice at any time under Settings → Privacy options; without consent, only non-personalised ads are requested. On any Android device you can also reset or delete your advertising ID, or turn off ad personalisation, under Settings → Google → Ads. Google's own handling of this data is described in its privacy policy at policies.google.com/technologies/ads.
What we deliberately do not collect
- No account, email address, phone number, or password — the app has no sign-in at all.
- No analytics, crash-reporting, attribution, or tracking SDKs are bundled in the app. The only third-party SDK is Google AdMob, in the Android app, for the advertisement described above; the iOS app currently contains no advertising.
- No contacts, photos, calendar, location, or health data.
- No payment information — the app does not take payments.
3. Who processes your data
We use a small number of third-party processors to deliver the translation itself. They act on our instructions and are bound by their own contractual and privacy obligations. We do not sell or rent your data to anyone, and we do not share your audio, transcripts or translations with data brokers or advertisers, or use them for any purpose other than delivering the translation you asked for. The advertising SDK in the Android app never receives any of that; what it does receive is set out in section 2.
| Processor | What it receives | Why |
|---|---|---|
| Google (AdMob — Android app only) | Your advertising ID, IP address, basic device and app information, and your interaction with the advertisement. Never your audio, transcripts or translations. | Serves and measures the advertisement shown when the app opens. Personalised only with your consent where consent is required; otherwise non-personalised. |
| Microsoft Azure (Azure Container Apps, Australia East region) | Hosts our server. Receives your audio stream, transcripts, and connection metadata. | Runs the application that coordinates the whole session. |
| ElevenLabs | Your microphone audio, the transcribed text, and the translated text. | Speech-to-text, and text-to-speech for the translated voice. In our ElevenLabs Agents configuration, ElevenLabs also performs the translation itself. |
| Our language-model provider: Microsoft Azure OpenAI (Azure AI Foundry), processed in Microsoft's Australia East region. It is used only when a session runs on our cascaded pipeline; in the default ElevenLabs Agents configuration the translation is produced by ElevenLabs, as described above. | The transcribed text of what was said, plus the source and target language. | Produces the translation. Not used when the ElevenLabs Agents engine is enabled. |
| Twilio (Network Traversal Service) | Your device's network address, and — when a direct connection is not possible — the encrypted audio packets, which pass through as relayed traffic. | Supplies the STUN/TURN relay that lets the audio connection form across mobile networks and firewalls. |
| Google (public STUN server, fallback only) | Your device's network address, if the Twilio relay is unavailable. | Fallback network-address discovery. No audio or transcript is sent to it. |
| Content delivery networks (unpkg, Google Fonts) | Your device's network address and browser user-agent when the app interface loads. | Serve the interface's fonts and JavaScript libraries. No audio or transcript is sent to them. |
These providers operate in multiple countries. If you are outside the country where a provider processes data, your information will be transferred internationally. For the deployed service that means Australia (Microsoft Azure, Australia East, which hosts the service and the language model) and the United States (ElevenLabs). We are an Australian company, and overseas disclosure is made under Australian Privacy Principle 8.
4. What is kept, where, and for how long
| Where | What is held | How long |
|---|---|---|
| Your device (iOS) | Session transcripts, speaker names, topic, languages, duration, timestamps — stored in the app's private SwiftData database. | Until you delete the app. See section 5. |
| Your device (Android) | The same session history, stored in the app's private database. | Until you delete the app or clear the app's storage. See section 5. |
| Our server | Transcript text and session metadata, held in the memory of the running process. Audio is streamed through and is not written to disk or to any database. | For as long as that server process is running. It is discarded whenever the service restarts, is redeployed, or scales down. Independently of restarts, a session is removed from server memory 24 hours after it ends, and a session that was created but never joined is removed 24 hours after it was created. Nothing about a session survives on the server beyond that. |
| Server logs | Timestamps, session codes, connection and error events. Not transcript text. | 30 days (Azure Log Analytics workspace retention). |
| Our processors | Whatever each receives per section 3, under that provider's own retention policy. | Each provider keeps what it receives under its own published retention terms. We have not enabled zero-retention options: ElevenLabs retains conversation records within our account under its terms, and Microsoft may retain Azure OpenAI inputs for abuse monitoring under its published terms. |
Please be aware: the app has no sign-in. A session's server-side copy is protected by its session code, which is long and randomly generated, and the app only ever lists sessions created on your own device. But the code is the only thing protecting that copy: anyone you share a session code with can view that session's transcript for as long as the server process is running. Share session codes only with people you intend to include in the conversation.
5. Your on-device history and how to clear it
Your conversation history lives in the app's own private storage on your device. It is not backed up to any account of ours, and we cannot read it.
- iOS and watchOS: deleting the app removes its database and everything in it. This version of the app has no clear-history control of its own; deleting the app is the only way to remove it.
- Android: Settings → Apps → Vocare Translate → Storage → Clear storage, or uninstall the app.
- If your device backs up app data (iCloud Backup, Google Backup), a copy of that history may be included in the backup, under Apple's or Google's own privacy terms.
To have the server-side copy of a session removed sooner than section 4 describes, contact us at the address in section 9 with the session code.
6. Children
Vocare Translate is not directed at children and we do not knowingly collect personal information from children under 13 (or the equivalent age of digital consent where you live). The app has no accounts, no user-generated content shared with others, and no social features. The Android app shows advertising from Google AdMob; the app is not directed at children and we do not request advertising aimed at children. If you believe a child has used the app and you would like their data removed, contact us and we will help.
7. Legal basis for processing
Where the GDPR, UK GDPR, or a similar law applies:
- Performance of a contract (Art. 6(1)(b)) — processing your audio and transcripts is what delivers the translation service you have asked for.
- Consent (Art. 6(1)(a)) — you grant microphone access explicitly through your operating system, and you may withdraw it at any time in your device settings, which stops all audio processing.
- Legitimate interests (Art. 6(1)(f)) — keeping the service secure, available and free of abuse, using the operational logs described in section 2.
Spoken conversation can contain special-category information (Art. 9) if you choose to discuss, for example, health matters. We do not seek it out and we do not analyse transcripts for any purpose beyond producing the translation, but you should keep it in mind when deciding what to say.
Australian users: we handle personal information in accordance with the Privacy Act 1988 (Cth) and the Australian Privacy Principles.
For advertising in the Android app, non-personalised advertising is served on the basis of our legitimate interest in funding a free app, and personalised advertising only on the basis of your consent where the law requires it.
8. Your rights
Depending on where you live, you may have the right to access, correct, delete, or export your personal information, to object to or restrict its processing, and to complain to your data protection authority. Because the app holds no account, we cannot look you up by name — a request about a specific session needs its session code. We do not sell personal information. In the Android app, the advertising SDK's use of your advertising ID for personalised advertising may amount to "sharing" for cross-context behavioural advertising as that term is used in California law; you can opt out by declining personalised ads under Settings → Privacy options, or by turning off ad personalisation in your device's Google settings.
9. Contact
- Privacy enquiries
- info@vocare.cx
10. Changes to this policy
If we change how the app handles your data we will update this page and change the date at the top. Material changes will also be noted in the app's release notes.