Authoring AI 2.0 brings Next Gen Narration to your presentations with richer, more natural audio and control over how your script is delivered. The Authoring AI 2.0 library sits alongside Authoring AI 1.0. For more information, see add audio narration to your slides.
See how it looks
Key features
- Richer, more natural audio, especially on longer narration and when reading numbers and dates
- Emotional context tags to direct how a line is delivered
- Phonemes in the pronunciation dictionary that work in all languages
- Inline IPA notation directly in your script
- Multilingual voices that can narrate content in any language
- Custom voices move to the new engine automatically and get the same capabilities
You need this to succeed
- Plan: All plans with Authoring AI
- Roles & Privileges: Presentation Author
- Folder Permissions: Author for at least one folder.
Do this step by step
Open the presentation editor
- To open the presentation in edit mode, click Content in the top menu and select My Content.
- Search and filter your Presentations to locate the presentation you'd like to edit.
- Click the three dot button on the presentation tile and select Edit.
Generate audio with an Authoring AI 2.0 voice
The generation workflow is the same as before. The only difference is the voice library you choose.
- Select the slide you want to generate audio for.
- In the Slide properties panel on the right side, click the Notes tab.
- Enter your script into the text box and click Save.
- Click Generate narration above the notes field.
- In the Voice library selector, select Authoring AI 2.0.
- Select a language and choose your preferred AI voice from the library. To hear a sample, click the play icon.
- If you'd like to ensure specific words and terms are pronounced as your organization prefers, check the Pronunciation dictionary box.
- Click Generate and save.
-
Click the play icon to listen to the narration.
Note: Authoring AI 2.0 launches with English voices. More languages will be added over time. All Authoring AI 2.0 voices are multilingual, so they can narrate content in any language, but they may keep their native accent.
Tip: To generate audio for all slides with notes at once, click Narration at the top of the page, select Generate narration, and choose an Authoring AI 2.0 voice.
Add emotional context tags to your script
Emotional context tags let you direct how a line is delivered, like giving a note to a voice actor. The narrator performs the emotion, not just the words.
- In the Notes tab, open the script for your slide.
- Add a tag in square brackets where you want the effect. Tags can go anywhere in a sentence, not only at the start. Tag categories include:
- Emotions: Shape the feeling of the delivery, for example [curious], [crying], [mischievously].
- Delivery direction: Control how something is said, for example [whispers], [shouts].
- Human reactions: Add natural texture, for example [laughs], [clears throat], [sighs].
- Click Save and generate the narration with an Authoring AI 2.0 voice.
Example: [clears throat] Let me walk you through what happened. And I have to say, [laughs] it wasn't what we expected.
Control pauses and pacing
- Use [pause], [short pause], or [long pause] to add pauses.
-
Commas, semicolons, colons, periods, and hard returns control pauses and pacing more consistently.
Note: The [break] tag doesn't work with Authoring AI 2.0 voices. It still works with Authoring AI 1.0 voices.
| Category | Tag | Example |
|---|---|---|
| Emotion | [proud] → [startled] | [proud] I’ve been cooking for over a decade. I know how to boil an egg. [startled] What’s that burning smell? |
| [smug] | [smug] Easy. It’s Saturn, everyone knows that. | |
| [triumphant] | [triumphant] YES! Saturn! We won! | |
| Delivery & volume | [whispers] | [whispers] Don’t move, it’s right behind you. |
| [shouts] | [shouts] Everybody evacuate the building, now! | |
| [low, threatening] | [low, threatening] You really shouldn’t have come here. | |
| Pacing | [slowly] | [slowly] And the winner is... |
| [rushed] | [rushed] Sorry, I’m late, the train broke down, I ran the whole way. | |
| [drawn out] | [drawn out] Nooo way. | |
| Human reactions | [laughs] | [laughs] You actually fell for that? |
| [sighs] | [sighs] Fine. I’ll do the dishes. | |
| [gasps] | [gasps] Is that a real diamond? | |
| Accent & character | [British accent] | [British accent] Fancy a cup of tea? |
| [Australian accent] | [Australian accent] No worries, mate, we’ll sort it out. | |
| [pirate voice] | [pirate voice] Hoist the sails and pass the rum. | |
| Sound effects | [thunder rumbling] | [thunder rumbling] It’s getting closer. |
| [door creaking] | [door creaking] Hello? Is anyone home? | |
| [clapping] | [clapping] Thank you, thank you, you’re too kind. | |
| Custom (natural language) | [speeding up, like a sports commentator] | [speeding up, like a sports commentator] He’s going for it, he’s reaching, he’s almost there... |
| [barely containing excitement] | [barely containing excitement] The intern. | |
| [sudden, crushed disappointment] | [sudden, crushed disappointment] Someone from accounts got there first. |
Control pronunciation with phonemes
With Authoring AI 2.0, phonemes work in every language, not only English. You can add them in two ways:
- Pronunciation dictionary: Add phonemes to the pronunciation dictionary for any language. Aliases continue to work in all languages.
- Inline International Phonetic Alphabet (IPA) in your script: Replace a word in your script with its IPA notation, directly in the sentence. You don't need a dictionary entry.
Read about the International Phonetic Alphabet for details.
Use custom voices with Authoring AI 2.0
Custom voices (voices you've uploaded and cloned) move to the new engine automatically. They get the same capabilities as Authoring AI 2.0 voices: better audio quality, emotional context tags, and expanded phoneme support. You don't need to do anything.
Authoring AI 1.0 voices remain available if you prefer them. You can switch libraries at any time in the voice library selector.