Summary: By default, Vocena processes sound and text locally on your device. Content is sent to third-party providers only when you choose to use cloud recognition, online translation, or cloud AI (LLM translation, AI replies, web search, polish, and the scene assistant); on iPhone and iPad, the app asks for your consent before it sends anything for the first time. Local mode blocks uploads and cloud processing, but models and system language packs can still be downloaded. Our servers do not store your recordings or text. We do not sell your data, and we do not use your content to train models.
1. Data controller
Vocena is developed and operated by Whyeung Digital LLC (registered in Wyoming, USA; “we” or “us”). We are the data controller for the processing described in this policy.
2. On-device processing
The following processing happens entirely on your device, and the related content is not transmitted to our servers or to any third party’s servers:
- Live captions and recording transcription: when you choose Apple speech recognition or the Qwen local model.
- Translation: when you choose Apple Translation. Apple processes it on the device, and it is subject to Apple’s own privacy policy.
- AI replies and polish: when you use Apple Intelligence on the device. Apple processes it without going through our servers, and it is subject to Apple’s own privacy policy.
- Library: recording files, transcripts, translations, polish results, and scene settings are stored only in the app’s data folder on this device by default. We hold no copy of your library and have no technical means to read it.
- App settings: language, scene, and interface preferences are all stored on this device.
3. Cloud processing and data sent
The app sends content only when you choose to use the features below; AI replies and polish that use Apple Intelligence are processed on the device (see section 2). You can turn each feature off separately. When Local mode is on, the app stops uploads and cloud processing, but models and system language packs can still be downloaded.
| Feature | What is sent | Recipient | Purpose |
|---|---|---|---|
| Cloud recognition | Recorded audio segments from the microphone and system audio | Alibaba Cloud (Tongyi Qwen) | Turn speech into text |
| Online translation | Caption text | NetEase Youdao, DeepL | Produce translations |
| LLM translation | Caption text | OpenAI, Google | Produce translations |
| AI replies | Caption text within the AI key’s reading range, the lines you select, and the questions and instructions you type | Same as above | Generate replies |
| Web search | The same as for AI replies; the provider searches the web based on it | Same as above | Add information that is not in the captions |
| Polish | The text to be polished, and the wording and layout rules you set | Same as above | Rewrite text |
| Scene assistant | Your description of a scene or the change you ask for, and the current scene’s settings | Same as above | Create or change scenes |
| Model downloads | None of your content; only requests for model files | Hugging Face, hf-mirror.com, ModelScope | Get local recognition models |
| Export to Notion (Mac only) | The content you choose to export | Notion | Write to your own Notion page |
| Account and payments | Email, verification codes, account and device identifiers, purchase and subscription status | Our own servers and Stripe; also Apple if you use Sign in with Apple | Sign-in and payment |
| Feedback | What you write in the feedback form, an optional reply email, and diagnostics if you choose to include them | Our own servers | Handle the problems you report |
Consent before sending: On iPhone and iPad, before the app first sends your data to the providers of cloud recognition, cloud AI, or online translation, it explains what will be sent and to whom, and asks for your consent. If you choose “Don’t Allow”, the app does not send it. You can turn off any of these at any time in Settings › General › Data Sharing; once one is off, the app stops sending and asks again the next time it is needed.
Link to your account: Cloud recognition, LLM translation, AI replies, web search, polish, and the scene assistant require a subscription. These requests carry your Vocena account token, so they are linked to your account; we record usage per account to calculate your quota and prevent abuse. Audio for cloud recognition may be relayed to Alibaba Cloud through our servers, or sent directly from your device to Alibaba Cloud after our servers issue a temporary credential. Text for online translation is sent directly from your device to NetEase Youdao or DeepL, without account information.
We will not: sell your audio or text; use your content to train our models, or provide it to any third party for training; upload any recordings or transcripts when you have not turned on a cloud feature.
About Local mode: When Local mode is on, the app stops uploads and cloud processing and no longer sends your recordings or text. Models and system language packs can still be downloaded.
4. Third-party services
We use third-party services only to the extent needed to provide the features you turn on. Each provider has its own privacy policy and is bound by its own terms when it processes the content you send:
- Apple — system speech recognition, system translation, Apple Intelligence, Sign in with Apple, and App Store purchases and subscriptions.
- Alibaba Cloud — cloud speech recognition (Tongyi Qwen models).
- OpenAI — large language models for LLM translation, AI replies, web search, polish, and the scene assistant.
- Google — large language models for LLM translation, AI replies, web search, polish, and the scene assistant.
- NetEase Youdao, DeepL — online translation. Your device sends the text directly to them. They are third-party web translation services that process, and may retain, the text they receive under their own terms; we cannot control how they use it.
- Cloudflare — network delivery and security for the vocena.app website and our server endpoints; requests between the app and our servers pass through Cloudflare’s network. Cloudflare also provides the website’s visit statistics and the human verification on the account pages (see section 6).
- Stripe — website payment processing. Your card number goes directly to Stripe; we cannot see it and do not store it.
- Hugging Face, hf-mirror.com, ModelScope — download sources for local model files.
- Notion — receives the content you export, only when you connect your own Notion on Mac and choose to export.
International transfers
We are a US company, and our servers and third-party providers are located in different countries and regions. When you turn on cloud recognition, audio is transferred to Alibaba Cloud servers in Singapore or mainland China, with the region assigned by our servers based on your network; when you use online translation, text is transferred to the servers of NetEase Youdao (China) or DeepL (Germany); when you use OpenAI or Google models, text is transferred to that provider’s servers, usually located in the United States. By turning on these features, you agree to these international transfers. If you do not want content to leave your device, use on-device recognition and on-device translation, or simply turn on Local mode.
5. Account and payment information
Free features do not require an account. You need to sign in only to buy a lifetime purchase or a subscription. We then keep:
- Your email address, for sign-in, verification codes, and transaction notices. When you use Sign in with Apple, we do not request your name or email from Apple; we identify you by the account identifier that Apple provides;
- Your account identifier, a random device identifier generated by the app, and the device type. They are reported at sign-in to bind the sign-in session to this device;
- Your purchase and subscription status and remaining quota percentage. When you buy through the App Store, the app sends us the transaction record signed by Apple so that the purchase is recorded on your account;
- Usage records for cloud features (which feature was used, audio duration, and amount of text processed), used to calculate your quota and prevent abuse; they do not include audio or text content;
- The website sign-in session, kept in an HttpOnly, Secure cookie and not written to the browser’s local storage.
We do not store your card number, CVV, or full billing details. Website payments are processed by Stripe, and in-app purchases on iOS are processed by Apple.
6. Website
- Network and security: vocena.app is served through Cloudflare. When you visit the website, Cloudflare processes your IP address and the technical information your browser sends, to deliver pages and defend against attacks.
- Mac update checks: Every few hours, the Mac app asks vocena.app whether a new version is available. Each check includes the app version, build number, macOS version, and chip type. Our server log keeps this with your IP address for 14 days, and we use it only to count how many Macs run each version. It is not linked to your account.
- Visit statistics: We use Cloudflare Web Analytics to measure visits to the website, such as the pages visited, the referring website, browser and device type, and country or region. It does not use cookies, and we only look at aggregate data.
- Human verification: Forms on the account pages, such as sign-in and registration, use Cloudflare Turnstile. It checks technical signals from your browser and device to detect automated abuse.
- Language preference: The website remembers the language you choose in your browser. This setting is stored only in your browser.
7. System permissions
- Microphone — records you and the sound around you, and is used only after you start it yourself.
- System audio recording — records the sound your device is playing (iPhone and iPad require iOS 27 or iPadOS 27), and likewise works only while you have it on.
- Speech recognition (iPhone, iPad) — required when you use Apple speech recognition.
- Notifications (optional) — status alerts, such as a finished transcription.
The app never starts recording on its own and does not keep listening in the background. While transcription is running, the menu bar icon on Mac, and the Lock Screen and Dynamic Island on iPhone, show its current status.
8. Retention
- Your recordings and text: stored on your device for as long as you decide. Once you delete them, they no longer exist, and we hold no copy.
- Content of cloud requests sent to third parties: handling ends once the request is forwarded; we do not write it to disk, index it, or keep any copy. Retention on the provider’s side is governed by each provider’s policy.
- Feedback: we keep the feedback you submit, the reply email, and diagnostics for up to 30 days. Diagnostics contain only the app and system versions, the app’s running state, and recent app events; they do not contain recordings or caption text.
- Account information (including device identifiers and usage records): kept while your account exists; after you delete your account, we delete it within 30 days, except for transaction records that the law requires us to keep.
- Payment records: kept for the period required by tax and accounting regulations.
9. Access and deletion
- Recordings and text: delete them in the app’s library. Deleted items first move to the trash; once you delete them permanently from the trash, the audio files and text records on this device are removed.
- All data on this device: uninstall the app and delete its data folder.
- Account: in the app on iPhone or iPad, open Settings › Account and choose Delete account. You can also delete it on the website’s account page, or email [email protected] to request deletion. Deleting your account removes your Vocena account, cloud access, and account tokens; recordings, captions, and the library on your device are not affected. If you use Sign in with Apple, we also revoke Vocena’s Sign in with Apple authorization when you delete your account. Deleting your account does not cancel an App Store subscription; cancel it in Apple’s subscription settings.
10. Your rights
To the extent permitted by applicable law, you have the right to request access to, correction of, and deletion of the personal information we hold about you, to request restriction of or object to certain processing, and to obtain a portable copy. You may also withdraw any consent you have given at any time; withdrawal does not affect processing completed before it. On iPhone and iPad, you can withdraw your consent for cloud recognition, cloud AI, and online translation in Settings › General › Data Sharing.
To make a request, contact [email protected]. We will respond within 30 days of receiving your request. If you believe our processing is improper, you have the right to lodge a complaint with your local data protection authority.
11. Children
Vocena is not designed for children. You must be at least 13 years old (or the higher age required by law where you live) to register a Vocena account. We do not knowingly collect personal information from children below that age. If you learn that a child has provided us with personal information, please contact us and we will delete it.
12. Lawful recording
Countries and regions have different rules on recording, and some require the consent of all participants before a call or meeting is recorded. You are responsible for deciding whether you have the right to record and whether you should inform others beforehand. Vocena is only a tool and does not obtain any consent on your behalf.
13. Changes to this policy
We may update this policy as features change. For material changes, we will show the new update date on this page and notify signed-in users through an in-app notice or by email. By continuing to use Vocena after the policy is updated, you accept the updated policy.
14. Contact
Privacy questions: [email protected]
For other questions, see the Support page.
Whyeung Digital LLC · Wyoming, USA
This policy is governed by the laws of the State of Wyoming, USA.