Apple has launched beta testing for its new Audio Intelligence technology on the Apple Watch, making the features available to developers in the iOS 27.3 update after their initial announcement at a hardware event in September. The suite introduces four opt-in artificial intelligence capabilities designed for Apple Watch Series 12 devices, including sound recognition, song identification via Shazam, Live Rewind, and Siri Recaps. These tools aim to assist users by detecting important environmental sounds, identifying background music, capturing missed dialogue snippets, and generating high-level conversation summaries.
The conversational tools, Live Rewind and Siri Recaps, process ambient audio while incorporating privacy and notification measures. Live Rewind allows users to double-press the digital crown to view a text snippet of the last 15 seconds of speech, accompanied by an audible chime and visual animation to signal recording. These snippets auto-delete within 30 seconds unless saved to the Siri app. Meanwhile, Siri Recaps generates concise conversation summaries with bullet points focusing on factual details like dates and deadlines, omitting speaker identification and raw audio retention. Users can review these summaries and export them to the Notes or Journal apps.
Why it matters: This rollout brings ambient audio AI capabilities to wearables while introducing on-device notification mechanisms to address surrounding privacy concerns.