Privacy and Architecture

Your voice
does not pass through us.

Voiced uses a direct-audio, ZDR text-routing architecture: audio goes from your Mac straight to the ASR provider; once it becomes text, Voiced servers briefly route it to the polish model. That handling is ZDR, not stored, and never used for training.

Data flow

Direct audio, ZDR text routing.

Here is the path for a full "hotkey → text at cursor" flow. The Voiced server is not in the audio path; text passes through briefly so it can be sent to the polish model.

AudioTextControl / subscription

CLIENT

Your Mac

Voiced.app · local

Audio, dictionaries, and history stay local by default

VENDOR · ASR

Speech-to-text

Zero data retention (ZDR) · no training

Released after transcription

VOICED · SERVER

Text polish gateway

ZDR · no content storage · no training

Brief text routing and subscription checks

VENDOR · LLM

Polish

Zero data retention (ZDR) · no training

Returned after rewrite, not retained long term

Put simply: audio never enters Voiced servers; polish text is handled briefly and released after the request finishes.

Three guarantees

Specific by design,
with no hidden path.

ZDR and no training

Voiced selects ASR and Polish providers with zero-data-retention terms: they release audio and transcripts after processing, do not train on your content, and sign data-processing agreements (DPAs). Provider details are written in the full policy below.

Direct audio, ZDR text

Audio goes from your Mac directly to the ASR provider, never through Voiced servers. Text briefly passes through Voiced servers on the way to the polish model, then is released instead of stored.

UK company, Taiwan team

Voiced is operated by ASWAY LTD, registered in England and Wales, and built by a Taiwan-based team. We handle personal data under applicable GDPR / UK GDPR and Taiwan Personal Data Protection Act requirements.

Cloud and local

What leaves
your device.

Voiced is a real Mac app. Most data stays local. During transcription, audio goes directly to the ASR provider; during polish, text briefly passes through Voiced servers and the polish model, then is released. Account and subscription data are kept by Voiced servers.

Stays localAudio, transcripts, polish results, history, dictionaries, and settings are saved on your Mac by default, with optional Touch ID encryption.
Local
Sent to providersAudio goes directly to the ASR provider during transcription; text is sent to the polish model during polish. Processing is released after completion (ZDR).
ZDR
Sent to Voiced serversAccount and subscription data are used for login and access checks; polish text only passes through briefly on the way to the model and is not stored.
ZDR

Full privacy policy

In writing,
clause by clause.

Last updated: 20 May 2026

01Data controller

Voiced is operated by ASWAY LTD, a company registered in England and Wales under company number 16800650, which is the data controller for the personal data described in this policy. Voiced is built by a Taiwan-based team, and we handle personal data under applicable GDPR / UK GDPR and Taiwan Personal Data Protection Act requirements. “We,” “us,” and “our” refer to ASWAY LTD.

02Our basic position

Voiced uses a direct-audio, ZDR text-routing architecture: your device communicates directly with speech-recognition (ASR) providers, so audio does not pass through Voiced servers. Once audio becomes text, that text briefly passes through Voiced servers on the way to language-model providers for polish. This handling is ZDR: we do not store transcript or polish content, and we do not use your content to train models.

03Data we collect

To provide the subscription service, we keep a small amount of account and operational data:

Email — for sign-in and account identification.
Subscription status — monthly/annual plan and renewal date.
Usage counts — monthly transcription count and total seconds (no content), kept in anonymous form.
Crash logs — to help us fix issues; can be turned off in settings.

04Data we do not collect

Your history, raw audio, and dictionary contents stay on your device by default. Aside from the brief text routing needed for polish, we do not upload those contents to Voiced servers for storage.

05How we use your data

We use the data above only to: create and verify your account, maintain and bill your subscription, provide technical support, fix bugs and improve stability, and comply with legal obligations. We do not sell your personal data to third parties.

06Legal bases for processing

Under the UK GDPR, our bases for processing include: performance of our contract with you (providing the subscription), our legitimate interests in maintaining and improving the service, and legal obligations where required. Optional items such as crash logs rely on your consent, which you can withdraw at any time.

07Third-party providers

We use third-party speech-recognition and language-model providers to process your voice and text. Speech recognition is initiated directly by your device with the ASR provider; polish text is briefly routed by Voiced servers to the language-model provider under ZDR handling. Primary providers include Vertex AI and DeepInfra. They are selected for zero-data-retention and no-training terms, and sign data-processing agreements (DPAs). Billing is handled by a payment provider.

08International transfers

Some providers may be located outside the UK or the European Economic Area. Where such transfers occur, we rely on appropriate safeguards such as Standard Contractual Clauses (SCCs).

09Data retention

Account and subscription data is kept for the life of your account and deleted within a reasonable period after closure, or the minimum period required by law. Voiced servers do not store polish text; it is released after the request finishes. Voice and text on the provider side are retained briefly or released immediately per their policies.

10Security and local storage

History, dictionaries, settings, and audio are stored locally on your device, with optional Touch ID encryption; whether to sync via iCloud is your choice. We apply reasonable technical and organisational measures to the account data we hold.

11Your rights

Subject to applicable law, you have the right to access, correct, or delete your personal data, to restrict or object to certain processing, to data portability, and to withdraw consent previously given. To exercise these rights, contact us using the details below. You also have the right to complain to the UK Information Commissioner’s Office (ICO).

12Cookies

This website uses only the essential cookies needed for basic operation and the checkout flow. We do not use them for cross-site advertising tracking.

13Children

Voiced is not directed at children, and we do not knowingly collect personal data from children.

14Changes to this policy

We may update this policy from time to time. When we do, we will update the “Last updated” date on this page and give reasonable notice of material changes.

15Contact us

If you have any questions about this Privacy Policy, or wish to exercise your rights, reach us via the contact page.

Voiced is operated by ASWAY LTD, a company registered in England and Wales. Company number: 16800650.

Ready to try Voiced?

Join the Waitlist

Your voicedoes not pass through us.