With the first beta of watchOS 27.3, Apple is introducing new Audio Intelligence features it previewed for the Apple Watch Series 12 and Ultra 4 earlier this year. Powered by the S11 chip, Audio Intelligence includes Live Rewind, Siri Recap, Sound Recognition, and Shazam Music Recognition.
Sound Recognition and Shazam are already on the watch, but Live Rewind and Siri Recap are now available in the beta. Live Rewind can show you a transcript of the last 15 seconds of a conversation. If you miss something like a waiter giving you the daily specials at a restaurant, you can press twice on the Digital Crown to see what was said without having to ask the person to repeat themselves.
Unlike Live Rewind, Siri Recap works in the background. The Apple Watch listens to conversations throughout the day and generates a high-level summary that includes a title and key points. Apple says you can use Siri Recap to jog your memory later, allowing you to stay present in the moment. If Siri Recap is turned on, summaries are created automatically and auto-deleted after seven days unless they're saved. It does not include a full transcript of conversations.
Audio Intelligence features are all opt-in. Apple says Live Rewind only activates when the Digital Crown is pressed. Transcription doesn't continue beyond the 15-second snippet, and text is not saved unless you ask Siri about it or save it to the Siri app.
Siri Recap can be scheduled to run based on time, location, or both. You can set it to listen just while you're at work, or add a rule to keep it from activating at night. It can also be turned on and off manually through Control Center, and Siri Recap summaries can be viewed and deleted in the Siri app.
Apple has a long privacy explainer for Audio Intelligence. Audio recordings aren't created or stored, and no audio can be accessed by apps, the operating system, or Apple. Raw audio is never available because it goes into the Secure Exclave on the S11 chip. Apple describes the Secure Exclave as a hardware-isolated compartment that processes sensor data separately from the rest of the system.
No Audio Intelligence features identify or attribute speakers, so text won't say "Eric said..." or label speakers as "Speaker A" and "Speaker B." Siri Recaps filter potentially harmful content and sensitive information.
When using Live Rewind, there is an audible chime that plays from the Apple Watch speaker, even if it's in silent mode. A full-display animation plays on the watch when Live Rewind is used, and a microphone indicator alerts those nearby that it's been turned on.
Live Rewind and Sound Recognition are processed on-device unless a Live Rewind snippet is sent to Siri for a request or saved. Siri Recap summaries are created through Private Cloud Compute.
Along with an Apple Watch Series 12 or Ultra 4, Siri Recap and Live Rewind require an iPhone 16 or later with Apple Intelligence and Siri AI. The new features are in English only to begin with, and they're not available to users under 13. Parental approval is required for Child Accounts under 18.
We're expecting watchOS 27.3 to launch in December alongside iOS 27.3.
