Project AIRI logoProject AIRI
Manual

Captions, Subtitles & Speech Bubbles

Captions, Subtitles & Speech Bubbles

Whether you are watching a stream, playing a game with background music, or prefer reading while listening, AIRI provides a versatile subtitle and captioning system that works across all render surfaces.


1. Caption Display Surfaces

AIRI supports several distinct caption presentation modes:

ModeAppearanceBehavior
Floating Caption WindowTransparent floating barCan be dragged anywhere on your desktop, docked below your avatar, or pinned to your taskbar.
Inline Chat CaptionsEmbedded chat bubblesDisplayed directly inside the 3-column Chatbox stream with emotion badges.
Comic Speech BubblesHead-tethered manga bubbles(Live2D & VRM) Floats dynamically above the character's head, bobbing with head tilt and movement.
Dating Sim Dialogue HUDBottom visual novel barClassic visual novel text box with typewriter effects and actor nameplates.

2. Sentence-Sync Audio Highlighting

AIRI features Sentence-Sync Audio Playback:

  • As the synthesized TTS audio plays, the active phrase or sentence is highlighted in real time.
  • Clicking any highlighted phrase replays that specific audio segment.
  • Transcribed microphone input from the user is displayed in an alternating accent color, making multi-turn dialogues easy to follow.

3. Customizing Caption Appearance

Open Settings → Appearance → Captions to adjust:

  • Typography & Font Size: Scale text from compact (14px) to large broadcast sizes (28px).
  • Background Opacity & Blur: Choose from solid contrast, frosted glass blur, or completely transparent background.
  • Docking Lock: Lock the caption bar to follow the avatar Stage window when dragging it across screens.
  • Auto-Hide Timeout: Automatically fade out subtitles after 5–10 seconds of silence.