AI voiceover for Keynote: narrate your slides on a Mac
Keynote can play audio on any slide, so a narrated deck can run without you. Instead of recording yourself slide by slide, you can voice the text of each slide with an AI voice, and fixing one sentence means voicing one line again.
This guide shows how to make the audio with the free SuperClonica engine and how to add it in Keynote on a Mac, using the steps from Apple’s Keynote guide.
Voice the text of each slide
Write what the narrator says on each slide in short spoken sentences and keep each slide in its own text file. Try voices free on their pages with a sentence of 30 to 100 characters, no account needed. Then install the free engine with python -m pip install superclonica-cpu and voice each slide: superclonica --text-file slide-01.txt --voice F2 --lang en --normalize --output slide-01.mp3. --normalize keeps every slide at the same loudness.
Add the audio to a slide
In Keynote, drag the audio file from Finder onto the slide, or click the Insert Media button in the toolbar, click Choose Music and drag the file onto the slide. Audio added to a slide plays only while that slide is shown, and by default it starts when you click it; Keynote can also start it automatically when the slide appears. The file must be in a format QuickTime supports on your Mac; if Keynote refuses it, Apple suggests converting audio to an MPEG-4 file with the .m4a extension.
Music under the whole deck
Background music for the whole presentation goes into the soundtrack: in the Document sidebar, open the Audio tab and add songs, then choose Off, Play Once or Loop. Keep the music much quieter than the voice, or leave it off for a lecture.
Before you share it
To play the deck on another Mac, turn on Copy audio and movies into document in Keynote’s General settings, so the narration travels with the file. Play the slideshow from start to finish once and check that every slide waits for its narration.
Your own voice or an AI voice
Keynote can also record you: Insert Media, Record Audio for one slide, or Record in the Audio tab of the Document sidebar to narrate the whole presentation while you advance the slides. An AI voice is handy when you need clean, even narration, many languages or quick edits. You can also train your own voice once in the studio from a 5 to 10 second recording and let it read every new deck.