---
title: "How to Narrate a PowerPoint: AI Voice or Your Own"
description: "Decide between AI narration and recording your own voice for a PowerPoint deck, then use a per-slide script template to place the clips and set the timings."
canonical: "https://www.murmurtts.com/blog/narrate-powerpoint-presentation-ai-voice"
---
[Murmur](https://www.murmurtts.com/)/[Blog](https://www.murmurtts.com/blog)/Guide

Guide

# How to Narrate a PowerPoint: AI Voice or Your Own

Decide between AI narration and recording your own voice for a PowerPoint deck, then use a per-slide script template to place the clips and set the timings.

![](https://www.murmurtts.com/murmur-icon.webp)Murmur·October 22, 2026·10 min read[Download free (3-day Pro trial)](https://murmur-updates.tarunyadav9761.workers.dev/download/latest)[Buy $49](https://murmur-licenses.tarunyadav9761.workers.dev/checkout)

Free plan with Kokoro voices, scripts up to 1,500 characters · 3-day Pro trial, then $49 once

On this page

Decide the voice source before you write a lineWhere each approach breaks downA slide script template you can fill in per slideGet the clips into PowerPointSlide timings and the export caveatCaptions and accessibility for narrated decksFrequently asked questions

[Back to the journal](https://www.murmurtts.com/blog)

To narrate a PowerPoint, choose the voice source first, because that one decision shapes the whole workflow. Record your own voice when the deck is personal, persuasive, or presented live by you, and generate the narration with text to speech when the deck will be revised, reused, or narrated the same way across dozens of slides. Once the voice is decided, the rest is shared work: a short script per slide, one audio file per slide, and a slide timing that matches each clip.

This guide compares AI narration with recording your own voice, gives you a decision table and a reusable slide script template, then walks the finished clips into PowerPoint on a Mac. It is written for course modules, training decks, and product walkthroughs rather than a live keynote you will deliver from the stage, because those are the decks that get edited after the first recording.

## Decide the voice source before you write a line

The voice source decides your cost, your revision loop, and how consistent the deck sounds from the first slide to the last. A recorded voice can feel warmer and more credible, but every wording change becomes a new recording session in a quiet room. Generated narration is editable one slide at a time, though you have to review names, numbers, and pacing before you trust it. PowerPoint's own recorder sits in between, and it is the lowest-setup option when the audio only needs to live inside the file.

| Factor | Recording your own voice | AI narration (text to speech) | PowerPoint's built-in recorder |
| --- | --- | --- | --- |
| Setup | A microphone, a quiet room, and time for retakes. | A text-to-speech app on your machine plus a written script. | Record Slide Show in PowerPoint, with no extra app. |
| Revisions | One wording change can mean re-recording a whole section. | Edit the line, regenerate one clip, swap the file. | Re-record the slide, or clear the take and try again. |
| Consistency | Pace and tone drift across a long deck. | The same voice carries every slide. | Your own voice, so it stays recognizably yours. |
| Privacy | Audio never leaves your room. | Local generation keeps the script off a cloud service. | Audio is embedded inside the .pptx. |
| Best for | Personal, persuasive, presenter-led decks. | Courses, training, and explainers that will change. | One-off decks that only need to play in a Slide Show. |
| Watch out for | Stumbles, background noise, and retake time. | Review names, numbers, and pacing; not ideal for subtle emotion. | On a Mac, an exported video may not carry separately inserted audio. |

Read the table as a starting point, then apply the checklist below to the deck you actually have.

- Choose your own voice when the message is personal or persuasive, when you are presenting it live, or when a client expects to hear you.
- Choose AI narration when the deck will be revised, when one voice must carry dozens of slides without drift, or when you cannot record clean audio on demand.
- Choose PowerPoint's built-in recording when the audio only needs to live inside the .pptx and you want zero extra tools.
- Mix the two when the opening must be personal but the body is reference material that is likely to change.

## Where each approach breaks down

Honesty about the tradeoffs saves a rewrite. A recorded narrator stumbles, and a single fixed sentence in the middle of a nineteen-minute deck can mean re-recording a whole section. Background noise, a laptop fan, or a passing siren can force a retake you did not plan for. The upside is authenticity: your audience hears the person who knows the material, and subtle emotion lands in a way a synthetic voice often cannot.

Generated narration has the opposite profile. It is fast to change and perfectly repeatable, but it flattens anything that depends on a knowing emphasis or a pause that lands just right. Names, acronyms, and product terms need review, because a voice that mispronounces a client name on slide two undermines the rest of the deck. Treat the script as a small quality gate: catch the errors in text before you generate audio, not after.

PowerPoint's built-in recorder is convenient, and Microsoft's help notes that narration will not record during slide transitions, so let a transition finish before you speak. The bigger limitation appears at export. Microsoft states that media inserted in previous versions of PowerPoint (which is linked rather than embedded), QuickTime media, macros, and OLE or ActiveX controls will not be included in a video created by PowerPoint. That page documents PowerPoint for Windows, and on a Mac the handling of separately inserted audio can vary by version. Test the actual export before you promise a video file.

## A slide script template you can fill in per slide

A per-slide script is short on purpose. Two spoken sentences on a content slide, one on a title or section slide, and stop. Read the slide title, then say the single idea the audience should take away from it. If a slide holds a table, do not read the table aloud; say what the table shows and move on. The template below keeps each slide's narration, its pronunciation notes, and its target time in one block, so you can regenerate a single line without touching the rest of the deck.

> Slide [number], [slide title] Point: [the one idea this slide has to land] Say: [one or two spoken sentences that state the point, not the on-screen text] Numbers and names: [how each should be read aloud or pronounced] Time: [target seconds for this slide]
>
> Reusable per-slide narration template

Keep the template rules light but consistent:

- Lead each slide's line with the point, not with a transition word.
- Write numbers and symbols the way you want them spoken, so you write five million rather than 5M.
- Put pronunciation notes next to proper nouns so the voice does not guess.
- Keep one script file with a heading per slide, so regenerating slide seven does not disturb the others.
- Cut filler that only makes sense when a cursor is moving on screen.

Here is what two consecutive slides might look like in that file. The wording is written to be spoken, and each block maps to one audio file name.

> Slide 2, Spot the hazard Point: the trainee identifies the hazard before acting. Say: On this slide, look at the loading bay and name the hazard in the photo. Notice that the spill sits inside the marked walkway. Numbers and names: none. Time: 14 seconds Slide 3, Report it Point: the report goes to the shift supervisor first. Say: Reporting starts with the shift supervisor, not the safety office. The form shown here takes under a minute to complete. Numbers and names: the word supervisor is one word, not two. Time: 15 seconds
>
> Example per-slide script excerpt

That structure is the reason a scripted deck is easier to voice than a free-form essay. If you are narrating a class, the [course narration guide](https://www.murmurtts.com/blog/text-to-speech-for-course-narration) covers the writing side in more depth, and the [training video narration guide](https://www.murmurtts.com/blog/local-tts-for-training-videos) covers the wider production loop. If the deck is really a product walkthrough, the [product demo script template](https://www.murmurtts.com/blog/product-demo-script-template) gives you a beat-by-beat structure you can borrow.

## Get the clips into PowerPoint

With the script written, the file work is mechanical. Generate one clip per slide, then place each clip on its own slide.

1. Write one or two spoken lines per slide and keep them in a single script file with a heading per slide.
2. Generate one audio file per slide with the same voice, and name each file to match the slide number, such as 03-report-it.m4a.
3. On a Mac, select the slide and choose Insert > Audio > Audio from File. Older Mac versions label the menu Audio Browser and Audio from File.
4. On the Playback tab, set Start to Automatically, and tick Hide During Show so the speaker icon does not cover the slide.
5. On the Transitions tab, under Advance Slide, tick After and enter the number of seconds the slide should stay on screen, just under the clip length.
6. Preview with Slide Show > From Beginning, then share the .pptx or test a video export before you rely on it.

Formats matter. Microsoft lists MP3, WAV, M4A or AAC, AIFF, and CAF among the audio types PowerPoint for Mac accepts, and names .m4a with AAC as its recommended format; avoid Windows Media formats such as WMA. A filename like 03-report-it.m4a sorts in slide order in Finder and tells you at a glance which slide it belongs to.

Murmur is a local-first macOS voice studio for Apple Silicon Macs running macOS 15 or later. It generates text to speech on your machine, so a deck that contains unpublished product names, internal figures, or anything under an NDA never has to be uploaded to a cloud service, and there are no per-character fees. You choose one voice, or design a consistent narration voice once, then generate every slide clip with it. Murmur costs $49 one-time, has a free plan and a 3-day Pro trial, and includes a 7-day refund policy. If you work on Windows, need live or real-time streaming speech, or want several people editing one project in the cloud at once, a cloud service with those features is the better fit, and it cannot transcribe a recorded interview because it does not do speech to text. The [Mac text-to-speech workflow](https://www.murmurtts.com/blog/mac-text-to-speech-workflow) explains the script-to-audio handoff, and the [audio export guide](https://www.murmurtts.com/blog/wav-m4a-mp3-tts-voiceover-export) covers master and delivery files.

## Slide timings and the export caveat

Timing is what makes the deck advance in step with the voice. On the Transitions tab, under Advance Slide, tick After and type the number of seconds the slide should stay on screen. Set the value just under the length of that slide's clip. Microsoft notes that manual timings can also be used to trim the end of a recorded segment, so a slide that ends with a second of silence simply advances earlier instead of waiting. Repeat per slide, or use Apply to All for one uniform value.

Now the export. Microsoft's page on turning a presentation into a video lists the items that will not be included in a PowerPoint-created video: media inserted in previous versions (linked media is dropped), QuickTime media, macros, and OLE or ActiveX controls. That page documents PowerPoint for Windows, and separately inserted audio can behave differently on a Mac by version. The safe habit is to keep the .pptx as the deliverable that is guaranteed to play, or export a visual-only video and add the per-slide clips in a video editor where you control the mix. Either way, watch the finished export with the sound on before you send it.

## Captions and accessibility for narrated decks

If the narrated deck becomes a video with audio, plan the captions at the same time. The W3C explains that captions are required for prerecorded video with audio at WCAG Level A, and that automatically generated captions are not sufficient by themselves unless they are confirmed to be fully accurate, because a single wrong word such as not can flip the meaning. Your slide script is the natural starting point for a caption file, but review it against the finished audio rather than assuming the draft text matches the spoken clip. Murmur does not do speech to text, so the transcript comes from your script or a separate transcription step, not from the app.

Practical order: finish the audio, generate or type the captions, fix the words that matter (names, numbers, and any negation), and only then publish. If the deck includes meaningful on-screen action that captions cannot carry, add a short description as well.

## Frequently asked questions

## Sources

- [Microsoft Support: Record a slide show with narration and slide timings](https://support.microsoft.com/en-us/office/record-a-slide-show-with-narration-and-slide-timings-0b9502c6-5f6c-40ae-b1e7-e47d8741161c)Accessed 2026-10-14
- [Microsoft Support: Add or delete audio in your PowerPoint presentation](https://support.microsoft.com/en-us/office/add-or-delete-audio-in-your-powerpoint-presentation-c3b2a9fd-2547-41d9-9182-3dfaa58f1316)Accessed 2026-10-14
- [Microsoft Support: Turn your presentation into a video](https://support.microsoft.com/en-us/office/turn-your-presentation-into-a-video-c140551f-cb37-4818-b5d4-3e30815c3e83)Accessed 2026-10-14
- [W3C Web Accessibility Initiative: Captions/Subtitles](https://www.w3.org/WAI/media/av/captions/)Accessed 2026-10-14

## Narrate your deck with one consistent voice

Murmur is a local voice studio for Apple Silicon Macs running macOS 15 or later. It generates the per-slide audio you insert into PowerPoint, on your own machine, with one voice across the whole deck. The website edition is $49 one-time, with a free plan and a 3-day Pro trial, and a 7-day refund policy. Listen to the public samples before you buy.

[Download free (3-day Pro trial)](https://murmur-updates.tarunyadav9761.workers.dev/download/latest)[Buy Murmur · $49](https://murmur-licenses.tarunyadav9761.workers.dev/checkout)

macOS 15+ · Apple Silicon required · 7-day refund policy

Made for your Mac

## Less reading. More making.

[Download free (3-day Pro trial)](https://murmur-updates.tarunyadav9761.workers.dev/download/latest)[Buy $49](https://murmur-licenses.tarunyadav9761.workers.dev/checkout)[Explore Murmur](https://www.murmurtts.com/#product)