Audio Classifier uses your device's microphone to feed a continuous audio stream into an on-device machine-learning model (YAMNet, provided by Google MediaPipe). The model classifies the audio and returns confidence scores for sound categories in real time.
The raw audio stream is processed in memory and discarded immediately after each inference frame. It is never written to disk, never stored, and never transmitted to any server or third party.
The only data persisted to your device is:
All of this data is stored exclusively in your device's private app storage. It is never uploaded anywhere.
Cloud backup and device-transfer backup are explicitly disabled. Your session data cannot leave your device through Android's backup system.
This app contains no analytics SDKs, advertising SDKs, crash reporters, or any other third-party code that transmits data off-device.
This app does not collect any personal information from anyone, including children under 13.
If this policy changes, the updated version will be posted at this URL with a revised date. Continued use of the app after a change constitutes acceptance of the new policy.
Questions or concerns? Email: bluer-splats9r@icloud.com