Privacy Policy

Effective August 28, 2026 · Developer Working Model LLC

# 1.0Overview

Audiary is a text-to-speech app for iOS. Your privacy is important to us. This policy explains what data Audiary collects, how it is used, and your choices.

# 2.0Data Collection

Audiary does not collect, transmit, or store any personal data on external servers. All user content (text, audiaries, settings, and listening history) is stored locally on your device.

2.1What stays on your device

Text you add or import into Audiary
Playback settings (voice selection, speed, theme, word tracking preference)
Listening statistics (total time listened, completed audiaries)
Custom pronunciation entries
Purchase status

2.2Apple services

Audiary uses the following Apple-provided services:

In-App Purchases (StoreKit): Purchase transactions are processed entirely by Apple. Audiary receives only a confirmation of purchase status. We do not have access to your payment information.
Crash Reports (App Store Connect): If you have opted in to share analytics with app developers through your iOS Settings, Apple may share anonymized, aggregated crash data with us. This data contains no personally identifiable information. You can control this in Settings > Privacy & Security > Analytics & Improvements > Share with App Developers.
Apple system voices (AVSpeechSynthesizer): When you select an Enhanced or Default voice, speech is synthesized on-device using Apple's built-in speech engine. No text is sent to any server.

2.3Neural voice model download

If you choose the Neural voice tier, Audiary performs a one-time download (about 90 MB) of the on-device AI voice model files. These files are hosted on Hugging Face, a third-party model-hosting service. This download contains model files only — no personal data, text, or identifiers are sent to Hugging Face, and Audiary makes no further network requests to it once the files are in place. All speech synthesis then runs entirely on your device.

2.4Web article import

When you share a web page to Audiary, the article text is normally extracted right inside Safari from the page you are already viewing — no network request is made by Audiary. When you share a bare link from another app (Mail, Messages, a different browser), Audiary fetches that one web page directly from its publisher in order to extract the article. This request goes only to the site you chose to share, contains no identifiers beyond a standard page request, and nothing about it is sent to us or to any other service. The extracted text is stored only on your device.

2.5PDF and EPUB import, and scanned-text recognition

When you import a PDF or an EPUB book, Audiary reads it entirely on your device. If the PDF has no text layer — a scanned book, for example — Audiary recognizes the text from the page images using Apple's Vision framework, which also runs entirely on your device. No page images, recognized text, or anything else about your PDFs or books is ever transmitted anywhere.

2.6Speech synthesis

All speech synthesis happens entirely on your device:

Neural voices are generated on-device by a locally-stored AI model (Kokoro 82M), executed through Apple's Core ML framework. Your text is never sent to any server for inference.
Enhanced and Default voices use Apple's built-in AVSpeechSynthesizer, which also runs on-device.

In all cases, the text you read in Audiary stays on your device.

2.7Diagnostic reports

Audiary uses Apple's MetricKit framework to save crash and performance reports directly on your device. These reports are never transmitted automatically — they stay on your device unless you choose to send them. If the app misbehaves, you can email the stored reports to us from Settings → Email Diagnostics; the email is composed on your device and nothing is sent until you tap Send. Reports contain technical details only (such as crash stack traces, device model, and iOS version) and never include the text of your audiaries, your settings, or anything you have listened to. You can review the attachment before sending it, and deleting the app removes all stored reports.

2.8Data we do NOT collect

No analytics or usage tracking
No automatic crash reporting — diagnostic reports stay on your device unless you email them to us
No advertising identifiers
No location data
No contacts or calendar access
No microphone or camera access
No network requests to third-party services, other than the one-time neural voice model download and the article fetch described above — the fetch goes only to the web page you explicitly shared, and no personal data or text is ever sent anywhere
No accounts or sign-ins

# 3.0Data Sharing

Audiary does not share any data with third parties. We have no servers, no databases, and no data pipelines.

# 4.0Data Storage and Security

All data is stored in your device's local storage using Apple's SwiftData framework. If you use the Share Extension to import content, that content is stored in a shared app container on your device — it is never transmitted externally.

# 5.0Data Deletion

You can delete any audiary at any time from within the app. Uninstalling Audiary removes all associated data from your device.

# 6.0Children's Privacy

Audiary does not knowingly collect any data from children under 13. Since the app collects no personal data from any user, there is no children's data to protect.

# 7.0Changes to This Policy

We may update this policy from time to time. Changes will be posted on this page with an updated effective date.

# 8.0Contact

If you have questions about this privacy policy, contact us at [email protected].