About narration audio

Narration audio in Dive SOPs is not recorded audio. Instead, AI automatically generates narration by reading aloud the step description or subtitle text. There is no field to input text directly into the audio.

To change the narration content (wording), edit the description or subtitle text that the narration reads from. For description editing, see Step title and description editing. For subtitle editing, see Subtitle editing. To fix only pronunciation issues (misreadings of kanji, units, or technical terms), see Correct misreading of kanji and units.

How to configure audio settings

Before you start

The SOP must be in draft status.

For video steps

You can configure settings at two levels.

The default setting for the entire block (step block) is set from the speaker icon at the top of the step block. It applies to all steps within the block.
To change settings per step, turn on "Define individually" for each step and configure the narration target and mute. You can make these individual settings either in "Edit video" (or "Edit" on smartphones) or in the step list for each step.

In audio settings, you can choose the "Narration target" from "None / Description / Subtitles". To turn off the original video audio, use "Mute video audio" (Video audio is automatically muted when subtitles are being read aloud). For detailed differences in narration targets, see Subtitle and description narration feature.

For non-video steps

Configure from "Audio". The narration target is description only.

Muting the original video audio

When you don't want to play back ambient noise or conversations from filming, turn on "Mute video audio". You can set this as either a default for the entire step block or as an individual setting per step. The configuration location is the same as described in "How to configure audio settings" above.

Note that for steps using subtitle narration, the original video audio is automatically muted.

If you want to change multiple or all steps at once, see here.