Audio Tags

Audio tags let you insert expressive cues (like laughs, pauses, whispers, and more) directly into your episode script. When audio is generated, these tags are converted into natural-sounding vocal effects.

Audio tags appear as inline chips within speech blocks and are triggered with a / slash command.

Inserting an Audio Tag

1. Click into any speech block in the script editor.
2. Type / to open the audio tag picker.
3. Browse the categorized list or start typing to filter tags.
4. Select a tag using Enter or by clicking it.

The tag is inserted as a styled chip inline with your text. For example:

"I just found out we hit a million downloads [laughs]. I honestly can't believe it."

Available Preset Tags

The picker is organized into four categories:

Reactions

• laughs: A light chuckle or laugh
• sighs: An exhale expressing emotion
• gasps: A sharp intake of breath
• clears throat: A brief throat clear

Emotions

• excited: Energetic, upbeat tone
• nervous: Hesitant, uneasy delivery
• calm: Relaxed, steady voice
• frustrated: Tense, irritated tone
• sarcastic: Dry, ironic delivery

Delivery

• pauses: A brief silence
• hesitates: Stumbling, uncertain speech
• dramatic: Intense, theatrical delivery
• whispers: Soft, hushed voice

Actions

• insert audio: Opens the Insert Audio dialog so you can drop a recorded clip into the transcript at this point. See Inserting Audio Into the Transcript.

Custom Tags

Not seeing the right tag? You can create your own:

1. Type / to open the picker.
2. Scroll to the bottom of the list or type a tag name that doesn't match any preset.
3. Enter your custom tag text in the Custom tag input field.
4. Press Enter to insert it.

Custom tags follow the same rules as presets: they're converted to vocal effects during audio generation. Keep custom tags short and descriptive for best results (e.g., chuckles nervously, takes a deep breath).

Editing and Removing Tags

• Click a tag chip in the editor to re-open the picker and replace it with a different tag.
• Backspace or Delete over a tag chip to remove it, just like deleting any other character.

How Tags Work During Audio Generation

When you generate episode audio:

• Each [tag] in the script is converted to a sound effect directive.
• The AI voice renders the tag as a natural vocal expression blended into the surrounding speech.
• Tags do not appear in captions, transcripts, or video subtitles. They're stripped from all user-facing text outputs.

Any text wrapped in square brackets in a speech block is treated as an audio tag, not just tags inserted through the picker. If you need literal square brackets in your script for something else (a citation, a timestamp, a version number), that text will be treated as a tag and stripped the same way a real tag would be, so avoid square brackets for anything you want spoken or shown in captions.

Frequently Asked Questions

Do audio tags use extra credits?

No. Audio tags are part of the speech generation and don't consume additional credits beyond the normal audio generation cost.

Will audio tags show up in my captions or transcript?

No. Audio tags are automatically stripped from captions, SRT files, clip transcripts, and all other text outputs. They only affect the audio.

Can I add audio tags to music blocks?

No. Audio tags can only be inserted inside speech blocks. The slash command is scoped to speech blocks only. Make sure you've clicked into a speech block (not a music block) before typing /.

Why doesn't the / slash command open a picker?

The slash command only works inside speech blocks with Horizon voices. If the picker doesn't appear, check: (1) you're clicked into a speech block (not a music block), (2) your host uses a Horizon voice (not a Classic voice), (3) you've actually typed the / character.

What happens if I type a custom tag the AI doesn't understand?

The AI voice will do its best to interpret the tag. For best results, use clear, short descriptions of sounds or delivery styles. If a custom tag doesn't produce the effect you want, try rephrasing it or use one of the presets.

Why do I see "This is an invalid section" when I try to regenerate?

A speech section needs actual spoken words, not just tags. If a section contains only audio tags (including custom tags with punctuation, like `[laughs loudly!]`), regeneration is blocked until you add spoken text to that section.