Which entries return sound with the picture

Sound reaches these sheets in two forms. One entry states that generated videos include an audio track by default, so a length request returns picture and sound. Two others animate a performance from audio the buyer provides, which is the reverse arrangement. As of 2026-09-22.

Inclusion rule. Statements each page makes about an audio track in its own output. A voice or music feature described elsewhere on a vendor site is not recorded from these pages. Order. The generated case first, then the supplied case.

1A default audio track changes the deliverable

A fifteen second request on that row returns fifteen seconds of picture and fifteen of sound. The duration parameter does not say so, which is why the page keeps the audio statement in the same place as the range.

Neither ad surface in this register states whether an audio track has to be present, so the default is neither required nor forbidden. A piece that will be laid under a separate voice track needs its own audio handled rather than assumed absent.

2Supplied audio makes sound the input, not the output

On the two performance rows the audio is what the job is built from, and the length figure follows it. That is why their ceilings are measured in minutes while every prompt-driven row is measured in tens of seconds.

It also means the voice is settled before generation rather than after. For an advertisement with a script that is often the natural order, and for one built around a visual idea it is the wrong way round.

3Silence on the rest is a gap worth naming

Most entries say nothing about whether output carries sound. A pipeline that assumes silence and gets a track, or assumes a track and gets silence, discovers it at the edit rather than at the brief.

The register records the statements and not the absence of a track, because absence cannot be verified from a page that never raises the subject. It can only be measured on a delivered file.

• Whether sound comes with the picture
  Generated videos include an audio track by defaultxAI docs, video generation / recorded 2026-09-22
• Longest output stated for Character-3
  Max duration: 10mHedra, Character-3 model page / recorded 2026-09-22

4A generator's ad pipeline

For ad work around “Sound that arrives”, SceneMixer's pages explain how its ad projects are produced:

Typically 60–120 seconds, always 9:16 vertical, ending on a hook. SceneMixer's episode outlines are generated with that pacing, and setting a project to 9:16 makes every segment vertical (the default is 16:9).

SceneMixer, How to Make ReelShort-Style Dramas with AI — “How long is a ReelShort-style episode?” (read 2026-10-13)

Measured across every job that finished successfully on scenemixer.com in the thirty days to 20 September 2026: analysis of a whole book runs about 2 minutes 49 seconds at the median, one episode's storyboard 1 minute 54 seconds, one video segment 3 minutes 20 seconds, and compositing an episode 3 minutes 52 seconds.

SceneMixer, Turn a Novel into a Video Series with AI — “How long does it take to produce one episode?” (read 2026-10-13)

5Sources

Format requirements come from Google Ads Help, read 2026-09-12, and tool limits from each vendor page listed on published limits, read 2026-09-22. The rule against mixing the two is on method. Nearby: Wider than widescreen, The square option, Identifiers and ceilings.