Privacy Policy

Effective Date: October 2025
Last Updated: April 2026
Version: 1.0.2

Changes: Clarified data handling and crash reporting behavior.

App Name: Sound Watcher © WaveAutomaton / Dmytro Iuryn

Sound Watcher performs local, real-time sound recognition entirely on-device. It does not record, store, or transmit audio or personal data. The app may transmit limited technical data in the form of crash reports.

1. Data Collection

No audio or personal information leaves your device.
No analytics or remote logging are used.
Optional crash reports may be sent and can include technical information such as error messages and stack traces. These reports are not designed to contain personal data.

_{* The app may send anonymous crash reports to help diagnose and fix errors.
Reports are intended to include error messages and technical stack traces and are not designed to contain personal data.
No personal data, account data, or audio recordings are collected or transmitted.
Reports are processed on WaveAutomaton-operated servers and are used solely to improve app stability.}

2. Connectivity

TThe app works offline and makes no network requests except for optional crash reporting. External links use HTTPS.

3. Model Credits

Sound recognition uses Google TensorFlow YamNet trained on AudioSet.
© 2019 The TensorFlow Authors — Apache 2.0 License.
AudioSet © Google — CC BY 4.0 License.

3.1. Model Limitations

Sound recognition is based on a pre-trained model and may produce inaccurate or incomplete results.

4. Your Rights

You may request confirmation that no personal data is held about you. We do not collect personal data, but will respond to any privacy inquiries.

5. Children

Sound Watcher is not directed to children under 13 (or 16 in some regions) and collects no data from anyone.

6. Policy Updates & Contact

This policy may be updated to reflect legal requirements. The current version is always available in-app or on our website.