Add a voice-over to a PowerPoint, Google Slides or Keynote presentation
A presentation that plays on its own, in a waiting room, on a trade-show stand or sent by email, needs narration. A synthetic voice produces it slide by slide from your notes, and updates when a slide changes.
One audio file per slide
The right unit is not the whole deck but the slide. Each slide gets its own file of 300 to 600 characters, twenty to forty seconds, well under the 1,500-character limit per generation. The benefit is twofold: playback stops naturally when the slide changes, and you only regenerate the slide you edited. Character counting includes spaces and punctuation; a slide that talks too much shows up on the counter straight away.
Your speaker notes are already the script
The notes field in PowerPoint, Google Slides or Keynote holds what you would say out loud. Rewrite it for the ear: one idea per sentence, present-tense verbs, numbers spelled out when they matter (« fourteen percent » rather than « 14% » if the voice hesitates), a full stop between two ideas to mark the pause. Our guides on writing a script for an AI voice and the seven mistakes that make a voice sound robotic cover the settings that make the difference.
Inserting the audio in PowerPoint, Google Slides and Keynote
In PowerPoint: Insert, Audio, Audio on My PC, then in the Playback tab tick « Start automatically » and « Hide during show »; the icon disappears and the narration starts with the slide. In Google Slides: upload the MP3 to Google Drive, then Insert, Audio, and pick « Autoplay » in the format options. In Keynote: drag the file onto the slide, then set automatic start in the Audio inspector. In all three cases, export as MP3 for a light file or WAV if your editing software prefers it; both formats are included, see pricing.
Exporting to video
PowerPoint (File, Export, Create a Video) and Keynote (File, Export To, Movie) assemble slides and narration into an MP4 ready for YouTube, an intranet or a stand screen. Google Slides needs a screen recording or a third-party tool. For a more polished corporate video, with music, live footage and captions, see the corporate video voice-over page.
One deck, two languages
The same notes, translated, give you the French version in minutes: Paul, Oliver or Jane in English, Marie in French, and you deliver two decks without re-recording anything. Every voice reads the same text on the voices page, which makes it easy to pick a consistent pair. Available languages are listed in the FAQ.
A steady tone from start to finish
For a business audience, Paul Confident in English and Marie Neutral in French carry a long deck without tiring the ear. Keep the same voice and tone on every slide: consistency is what makes it feel like a presentation, not variety. If one slide calls for energy, a title or an announcement, Paul Cheerful makes the difference over ten seconds. How the engine works and where it stops is on the AI voice generator page.
Update without redoing everything
A figure changes, a slide is added: you regenerate that one file, in the same voice, and replace the audio on the slide. That is the decisive difference from a human recording, where the smallest fix means booking the same voice in the same conditions again. Audio stays in your history for 90 days; keep your files in the presentation folder.
What it costs
A twenty-slide deck is about 8,000 characters, which is €0.72 with the €9 pack of 100,000 characters, valid for twelve months. The free trial, 160 characters with no account, covers a first slide so you can check the voice on your own text. For whole series of modules, the e-learning voice-over page describes the same method at scale, and the price comparison puts that cost next to the other tools. Commercial use is included with paid credits, see usage rights.