Apple has introduced four opt-in "audio intelligence" features on its new smartwatches, using the devices' microphones to power sound and music recognition, a conversation recap tool, and a "Live Rewind" function that shows a transcript of the previous 15 seconds of environmental audio. Anticipating unease about an always-listening wearable, Apple has moved to stress that the features are built around extensive privacy and security infrastructure, processing most data on-device and never storing raw audio. The announcement highlights how AI-powered listening tools are becoming increasingly embedded in everyday devices.

Central to Apple's approach is the new S11 chip's "Secure Exclave", an isolated memory space inaccessible to the rest of the operating system, where sensor data is processed. Sound Recognition, which flags noises such as doorbells or a crying baby, never leaves the watch, while Shazam's music identification generates only a song "signature" that is deleted immediately after use rather than sending an audio file. The "Siri Recap" conversation feature detects speech without recording, encrypts audio for transfer to a paired iPhone via secure Bluetooth, transcribes and condenses it locally, screens it for harmful content, and deletes the raw audio at each stage before sending only a distilled, encrypted summary to Apple's Private Cloud Compute servers.