How to Enable and Use Live Captions on iPad for Real-Time Audio Transcriptions

Whether attending a video conference, watching uncaptioned educational videos, listening to podcasts, or having in-person conversations in noisy environments, following along with spoken dialogue can be challenging. In iPadOS, Apple includes Live Captions, a system-wide accessibility feature that generates real-time text transcriptions for any spoken audio playing on your device or spoken into the microphone. Powered by on-device machine learning, Live Captions transcribe audio completely offline without transmitting private recordings over the internet.

Taking advantage of the iPad's large display canvas, Live Captions can be positioned as a floating dock, expanded into a full-screen transcription ledger, or integrated directly into FaceTime calls.

Requirements and Device Compatibility

Live Captions utilise the Apple Neural Engine for real-time speech processing:

• Device Support: iPad models equipped with the A12 Bionic chip or later (including iPad 8th generation and newer, iPad Air 3rd generation and newer, iPad mini 5th generation and newer, and all iPad Pro models with Apple Silicon M-series or A12Z/A12X chips).
• Operating System: iPadOS 16, iPadOS 17, or iPadOS 18.
• Language Availability: Primary support in English (US and UK), with expanding language support across recent iPadOS updates.
• On-Device Privacy: All speech-to-text processing occurs locally on your iPad. Audio streams from calls, videos, and microphones never leave your hardware.

Enabling Live Captions in iPadOS Settings

Live Captions are housed in the Accessibility section of system settings:

1. Open the Settings app on your iPad.
2. Scroll down the left navigation panel and tap Accessibility.
3. Under the Hearing section on the right, select Live Captions (Beta).
4. Toggle the Live Captions switch to the On position.
5. A compact, floating Live Captions window immediately appears on your screen, indicating that the transcription engine is active.

Customising the Floating Live Captions Window

The Live Captions interface is designed to stay accessible without obstructing your active workflow:

• Repositioning: Touch and drag the top header bar of the caption window to place it anywhere along the perimeter of your display.
• Minimising to an Icon: Tap outside the window or tap the collapse arrow to shrink Live Captions into a small floating microphone bubble docked against the screen edge. Tap the bubble anytime to expand it back to full size.
• Expanding to Full Screen: Tap the expand icon in the caption toolbar to enlarge the window into a reading pane, making lecture transcripts easily readable from across a desk.
• Pausing Transcriptions: Tap the pause icon inside the floating controller to halt listening during breaks or confidential discussions.

Customising Text Appearance and Font Size

To ensure captions remain legible regardless of your eyesight or background wallpaper colour:

1. In Settings > Accessibility > Live Captions, tap Appearance.
2. Adjust the Text Size slider to enlarge or reduce transcription font dimensions.
3. Choose custom Text Colour and Background Colour options.
4. Adjust Idle Opacity to control how transparent the floating controller becomes when no audio is currently playing.

Using Live Captions in FaceTime and Video Calls

Live Captions provide enhanced speaker identification during video calls:

1. In Settings > Accessibility > Live Captions, ensure Live Captions in FaceTime is toggled on.
2. During an active FaceTime call, live dialogue appears in a dedicated overlay.
3. When multiple participants speak, Live Captions attributes dialogue to individual callers using their contact names, making group discussions easy to follow.
4. You can also combine Live Captions with Type to Speak, allowing non-verbal users to type responses that Siri speaks aloud to other call participants.

Transcribing Real-World In-Person Conversations

Beyond app audio and video playback, Live Captions can transcribe face-to-face dialogue:

1. Tap the floating Live Captions controller on screen.
2. Tap the Microphone icon in the controller toolbar to switch from device audio to the iPad's ambient microphone.
3. Place your iPad on a table between speakers.
4. Spoken conversation transcribes onto the screen in real time, making lectures, meetings, and doctor consultations fully accessible without requiring an internet connection.