Privacy Policy

Last updated: November 8, 2026

This Privacy Policy describes how Felix Thron ("we", "us", or "our") collects, uses, and protects your information when you use the SpeechPacer mobile application ("App").

1. Information We Collect

1.1 No Account, No Personal Profile

No account is required, and SpeechPacer never asks for your email address, phone number, or other contact details. You can optionally enter a name in Settings for the personal greeting in Practice; it stays on your device and is used only for that greeting. Purchases and your subscription status are handled by Apple's StoreKit on your device. We never receive your Apple Account details, payment information or transaction identifiers.

1.2 Speech and Audio Data

When you start a coaching session:

  • The microphone captures your speech for real-time pace analysis. SpeechPacer asks only for microphone access for this and does not request a separate speech-recognition permission
  • All speech processing happens entirely on your device: speech-to-text uses Apple's SpeechAnalyzer framework (iOS 26+), and pause and pitch analysis use models built into the App
  • Audio is never sent to any server - not ours, not third parties
  • Recording is on by default: audio is saved as an AAC file on your device so you can play your sessions back. You can turn recording off during onboarding or at any time in Settings -> Speech & recording -> Record Sessions. Recordings stay on your device until you delete the session or exercise result they belong to; a recording that several saved entries share is deleted together with the last of them

1.3 Session Data

During and after each coaching session, the App calculates and stores:

  • Speaking pace (words per minute) sampled every 3 seconds
  • Zone classifications (good, caution, warning) based on your target pace
  • Session statistics (duration, average WPM, pause frequency, alert count, session score)
  • Pause, pitch and clarity measurements
  • The transcript of your session (the recognized words and their timing) for playback and feedback
  • Exercise results and practice progress (for example streaks, levels and milestones)

Texts you save yourself for reading practice are also kept on your device. All session data is stored locally on your device via Apple's SwiftData framework. The App never sends your recordings or transcripts off your device. Session metrics and results are shared only with your own Apple Watch (section 1.5), and if you opt in to analytics, only coarse ranges derived from your practice, such as a duration range, a score band or a range for the number of live alerts, are included in the analytics described in section 3. This also covers which of our authored exercises, lessons, knowledge units and practice packs you open, complete or download; texts you save yourself, their titles and your answers are never included.

1.4 User Settings

The App stores your preferences locally:

  • Target speaking pace and zone thresholds
  • Speech recognition language
  • Alert preferences (visual, sound, haptic, Watch haptic)
  • Recording toggle
  • Analytics consent (opt-in) and crash-report toggle (on by default, switch off any time)
  • Notification preferences
  • Optional name for the personal greeting in Practice

Settings are shared between the app and its widgets via Apple's App Groups secure container on your device. If you use an Apple Watch, it receives the settings it needs from your iPhone as described in section 1.5.

1.5 Apple Watch Data

If you use SpeechPacer with an Apple Watch:

  • Live session metrics (WPM, zone, alerts) are sent to your Watch via Apple's WatchConnectivity framework
  • The settings the Watch needs (for example Watch haptics and your target pace) and your last session result are sent the same way, and the Watch keeps your last score so its SpeechPacer widget can show it; session controls you use on the Watch are sent back to your iPhone
  • iOS and watchOS carry this communication over Apple's encrypted connection between your paired devices (Bluetooth, or Wi-Fi or cellular when the Watch is out of Bluetooth range); the App does not send it to us or to any third party

1.6 Coach's Note Generation

On devices where Apple Intelligence is available, SpeechPacer uses Apple's on-device Foundation Models to generate a personalized coach note after a session, an automatic session title, a short summary of what you said, a brief topic note that is used only to spot recurring themes across your sessions, and short notes on your progress trend. Everything is created entirely on your device - your session data and transcript never leave your device for AI generation. The generated texts are stored locally with your sessions and are never transmitted to our servers or any third party. You can turn this off in Settings -> Coach with Personal Feedback or On-device AI; automatic session titles and summaries can also be switched off separately.

2. How We Use Your Information

We use collected information to:

  • Provide real-time speaking pace coaching
  • Display session history and progress trends
  • Send local notifications (weekly summaries, practice reminders, and a notice when a session reaches its two-hour limit)
  • Deliver haptic alerts to your paired Apple Watch
  • Display session data in widgets
  • Understand how the App is used in aggregate, and detect and fix crashes (see section 3)

3. Analytics and Crash Reporting

SpeechPacer uses Google Firebase Analytics and Firebase Crashlytics to understand how the App is used in aggregate and to detect and fix crashes. We run them in a privacy-hardened configuration:

  • No advertising identifier (IDFA) is collected; the AdSupport framework is not linked
  • No vendor identifier (IDFV) is collected
  • No App Tracking Transparency prompt is shown, and no cross-app or cross-site tracking takes place
  • No user identifier is set. All user properties are coarse categories or bucketed ranges (for example a target-WPM band, a session-count bracket or the practice language), never values that identify you. Events from one practice attempt also carry a random identifier that is created for that attempt only
  • Your audio, transcripts, and session content are never sent to Firebase

The data Firebase collects falls into seven categories, all declared in our App Store privacy label as "Not Linked to You" and "Not Used to Track You":

  • Device ID - a per-install Firebase identifier
  • Crash Data - diagnostic logs when the app crashes, and reports of non-fatal errors when a session or recording cannot start
  • Performance Data - technical metadata about app usage sessions that the Firebase SDKs use to group crash and stability information (for example a random session identifier, when the App was opened, the device model and the iOS version), sent while Crash Reports are on; after you opt in to analytics, also coarse timing ranges, such as how long the on-device coach note took to generate
  • Product Interaction - which screens and features are used, including coarse practice results such as a duration range or a score band; which of our authored exercises, lessons, knowledge units and practice packs you open, complete or download; and coarse ranges for how often live alerts fired and how often your pace returned to target afterwards
  • Purchase History - whether a SpeechPacer Premium purchase or restore succeeded, was cancelled, is pending or failed, and in which part of the App the offer was opened. Collected only after you opt in to analytics; no product, price, transaction or account identifiers are sent
  • Coarse Location - the approximate region, such as the country or city, that Google Analytics derives from your device's IP address. Collected only after you opt in to analytics; the App never reads your device's location
  • Other Diagnostic Data - technical diagnostics from the Firebase SDKs, such as installation and data-delivery status

Legal basis: analytics and crash reporting have different legal bases and different defaults.

  • Firebase Analytics (feature usage) - consent, opt-in: processing happens only after your consent under § 25(1) TDDDG together with Art. 6(1)(a) GDPR. Analytics is off by default and becomes active only after you explicitly opt in by turning it on in Settings. You can withdraw your consent at any time.
  • Firebase Crashlytics (crash reports) - legitimate interest, on by default: this processing is strictly necessary for the stability of the service under § 25(2) TDDDG and relies on our legitimate interest in detecting and fixing defects under Art. 6(1)(f) GDPR. Crash reports are on by default; you can disable them in Settings. This change takes effect on next launch.

How to withdraw consent: to withdraw or re-grant your consent to analytics, open the App and go to Settings -> App & privacy -> Analytics. To turn off crash reports, go to Settings -> App & privacy -> Crash Reports. Switching Analytics off stops analytics collection at once, clears the analytics data still stored on your device and resets the per-install analytics identifier. Switching Crash Reports off stops reports of non-fatal errors at once and stops sending crash reports and the related session metadata to Firebase from the next launch.

Recipient and international transfer: the processor is Google (Firebase). Data may be processed in the United States. This transfer is covered by the EU-US Data Privacy Framework, under which Google LLC is certified (a European Commission adequacy decision under Art. 45 GDPR); EU Standard Contractual Clauses (Art. 46 GDPR) apply as a fallback safeguard.

3.1 Apple System Services

Beyond Firebase, the App relies only on Apple system services:

  • Apple Speech Recognition - on-device SpeechAnalyzer (iOS 26+) for speech-to-text. On first use, Apple may download a language model for your selected language. All subsequent speech processing occurs entirely on your device
  • Apple Background Assets - downloadable practice audio is delivered from Apple's servers through Apple-hosted Background Assets. iOS manages these downloads, and Apple delivers them under its own terms; they contain only the App's own practice recordings, and the App sends none of your session data or recordings with them
  • Apple StoreKit - handles SpeechPacer Premium purchases, restores and subscription management. Your subscription status is verified on your device; Apple processes purchases under its own terms, and we receive neither your payment details nor your Apple Account information
  • Apple Notification Services - local notifications are scheduled and delivered by iOS on your device. No push notification server is involved

4. Data Storage and Security

  • Your session data, settings, and recordings are stored locally on your device using Apple's SwiftData framework and the App Groups secure container; audio recordings are saved as AAC files in the app's private Documents directory
  • The App never uploads this on-device data to us or to any cloud service. If you back up your device with iCloud Backup or to a computer, iOS may include your session data (including transcripts and the coach notes and summaries generated from them), saved reading texts and settings in that backup; audio recordings and downloaded practice audio are excluded from backups
  • Apart from the communication with your own Apple Watch (section 1.5), the only data the App sends off your device is the analytics and crash data described in section 3, which carries no name, account or contact details; purchases and downloads are handled by iOS with Apple (section 3.1)
  • Your data benefits from the device-level encryption provided by iOS, and analytics in transit to Firebase is protected by TLS

5. Data Retention and Deletion

  • Sessions: Stored indefinitely until you delete them. You can delete any saved session or exercise result in Settings -> App & privacy -> Manage saved data, with or without Premium; where the History view is available to you, you can also delete sessions there. Deleting a session permanently removes the session record, all associated data points and analysis, and its audio recording; a recording that another saved entry still uses is removed together with that entry
  • Saved reading texts: Stored on your device until you delete them in Settings -> App & privacy -> Manage saved data, which is always free, or in Practice -> Read aloud, or until you uninstall the App. Without an active subscription they stay on your device and are hidden in Practice, and you can still delete them in Settings
  • Practice progress (streaks, levels and milestones): Kept for the lifetime of the app installation and removed when you uninstall the App
  • Settings: Persist for the lifetime of the app installation. Return most of them to their defaults with Settings -> App & privacy -> Reset to Defaults, or reset all of them by deleting and reinstalling the app
  • Analytics and crash data: retained by Google/Firebase under the retention settings of our Firebase project (analytics event data for 2 months, user-level analytics data for 14 months after your last activity; crash reports up to 90 days). Switching Analytics off in Settings stops new collection
  • Complete data removal: Uninstalling SpeechPacer from your device removes the App's on-device data, including the shared App Groups container. Copies in existing iCloud or computer backups remain until you delete or replace those backups, and iOS may keep a technical Firebase installation entry in the device keychain, which a reinstall does not reuse

6. Children's Privacy

SpeechPacer does not knowingly collect personal information from children under 13. The App requires no account or contact details at any age, and the data described in sections 1 and 3 is handled in the same way for every user.

7. Your Rights

Under the GDPR (European Economic Area)

You have the right to:

  • Access (Art. 15) - your settings are visible within the App at any time, and Settings -> App & privacy -> Manage saved data lists your saved sessions and exercise results; the full details of past sessions are shown in the History view with SpeechPacer Premium
  • Rectification (Art. 16) and Erasure (Art. 17) - change your settings and optional name in Settings at any time; delete any saved session, exercise result or saved reading text in Settings -> App & privacy -> Manage saved data, free of charge; or uninstall the App to remove all local data, including saved reading texts and practice progress
  • Restriction (Art. 18) - disable recording, notifications, analytics, and Watch features independently via Settings
  • Portability (Art. 20) - session data export is not currently offered; all data remains on your device
  • Withdrawal of consent (Art. 7(3)) - analytics is processed only on the basis of your consent; withdraw it at any time in Settings -> App & privacy -> Analytics, without affecting the lawfulness of processing carried out before withdrawal
  • Objection (Art. 21) - object to the crash-report processing carried out under our legitimate interest by switching Crash Reports off in Settings -> App & privacy -> Crash Reports

Because SpeechPacer keeps your session data on your device and shares only the analytics and crash data described in section 3, which carries no name, account or contact details, most data-subject requests are fulfilled directly by the App's built-in controls.

Right to lodge a complaint (Art. 77 GDPR): you may contact the supervisory authority responsible for us:

Berliner Beauftragte für Datenschutz und Informationsfreiheit
Alt-Moabit 59-61, 10555 Berlin, Germany
mailbox@datenschutz-berlin.de

Under the CCPA (California)

SpeechPacer does not sell or share personal information. The analytics and crash data described in section 3 are not used to identify you.

8. Changes to This Privacy Policy

We may update this Privacy Policy from time to time. Changes will be reflected in the "Last updated" date shown at the top of this page. Continued use of the App after changes constitutes acceptance of the updated policy.

9. Contact

If you have questions about this Privacy Policy, contact us at:

Felix Thron
Email: felix@speechpacer.com

Our postal address is listed in the Legal Notice.