Lilt Teleprompter

Lilt is a teleprompter built for people who speak to camera. Instead of scrolling at a fixed speed, it listens to your voice and moves the script as you read. Pause to think, slow down when you need time, add an extra line, or restart a phrase. Lilt waits for you and continues exactly where you left off. A live front-camera preview sits behind your words, so your gaze stays on the lens and your delivery looks natural. When you are ready, record straight to a 1080p video with audio. Your takes stay on your device until you decide what to do with them. Note: Lilt is not free to use. The free download lets the app check whether your iPhone supports on-device speech recognition before you commit to paying. You can also practice with a selection of public domain texts. Built around your voice - On-device speech recognition paces the script as you speak, so there is no fixed scroll speed to fight. - The cursor stays in the upper third of the screen, keeping the next few words within reach at all times. - If you pause to gather a thought, the script holds its place instead of racing ahead. - You can scrub the script by hand at any time, and manual control never interrupts your recording. Made for reading to camera - Front-camera preview behind the script for a natural eye line. - Mirror mode for beam-splitter teleprompter rigs. - Preamble mode keeps the script still while you deliver a spoken slate, then releases when you begin. - Record 1080p video with audio, saved to a file the app keeps for you. - Record as many versions as you like, then mark the good takes. Speaks your language - Choose a recognition language per script from the languages your device can transcribe. - Recognition runs quickly and entirely on the device. No wi-fi connection needed while you speak*, no network latency to worry about. Private by design - No analytics, no tracking, no accounts. - Speech recognition is fully on-device, and nothing you say is sent anywhere. - Your takes are stored privately on your iPhone. Saving a recording to Photos is always your choice and permission to access your library is only requested when you export. * Data connection required in advance to download language recognition data