Microsoft adds Audio UI Tokens and Live Caption Styles to Fluent to cover hearing-impaired and neurodivergent users

Tech · 4 min read

The new Audio UI Tokens encode parameters such as cue priority, spatialization hints, voice gender-neutral defaults, and adaptive volume envelopes so designers can standardize how systems use sound for alerts, confirmations and ambient feedback. Microsoft documented recommended mappings between visual component states (error, success, disabled) and corresponding audio tokens to avoid conflicting cues and to support multi-sensory feedback for neurodivergent users.

Live Caption Styles provide tokenized typographic, color and layout presets for on-device caption rendering, encompassing size scales, speaker-label treatments, background translucency and reading-mode toggles. Fluent’s update also adds a ‘Caption-aware’ layout constraint that components can opt into so captions don't occlude interactive elements on constrained screens like mobile or compact overlays.

Microsoft is releasing developer previews for Windows SDK, a Teams plugin, and Xbox developer toolkit that includes sample integrations for Unity. The company says these additions aim to reduce ad-hoc audio and caption implementations that have historically produced inconsistent experiences and accessibility regressions across apps.