AI video with sound

AI video with sound: dialogue, effects and music

Ten of the 13 video models on this page make sound in the same run as the picture, so the clip arrives with its soundtrack. On all four Seedance models, FLUX 3 Video and MiniMax H3 (Hailuo 3.0), that sound includes dialogue. For a 5 s, 720p clip, the lowest price with sound in the table below is $0.65, with Seedance 2.0 Mini and Kling 3.0 Omni.

Updated

First month free on the Beginner plan, with 650 credits to try any model.

AI video with sound, made in Root

Video models that make sound

The same short clip with sound on and off, priced on Root in dollars. 100 credits is $1.

ModelThe soundCan you turn it off?ClipWith soundWithout sound
Seedance 2.5Dialogue, effects and musicYes5 s, 720p$1.99$1.99
Seedance 2.0Dialogue, effects and musicYes5 s, 720p$1.30$1.30
Seedance 2.0 FastDialogue, effects and musicYes5 s, 720p$1.04$1.04
Seedance 2.0 MiniDialogue, effects and musicYes5 s, 720p$0.65$0.65
Kling 3.0A soundtrackYes5 s, 720p$0.73$0.49
Kling 3.0 OmniA soundtrackYes5 s, 720p$0.65$0.49
Kling 2.6A soundtrackYes5 s, 1080p$0.81In the app
Veo 3.1A soundtrackNo4 s, 720p$1.84—
FLUX 3 VideoDialogue and other soundYes5 s, HD$0.98$0.98
MiniMax H3 (Hailuo 3.0)Dialogue, effects and musicNo5 s, 1080pabout $2.39—

Kling 2.5 Turbo, Runway Gen-4.5 and Luma Ray 3.2 make silent clips. Kling 2.6 makes sound at 1080p only, so its clip with sound is made, and priced, at 1080p. MiniMax H3 (Hailuo 3.0) is priced at its High quality, the default; at Turbo quality the same clip is about $0.33. “In the app” means the app shows that price before you run it. Prices marked “about” are for work that runs on Root's own machines: the price shown before the run is an estimate, and the bill is set by how long the run really takes.

What sound costs

  • Nothing extra on all four Seedance models and FLUX 3 Video: the clip costs the same with sound or without.
  • More on Kling 3.0 and Kling 3.0 Omni. A 5 s, 720p Kling 3.0 clip is $0.73 with sound and $0.49 without.
  • Always included on Veo 3.1 and MiniMax H3 (Hailuo 3.0), which have no silent setting.

How to write sound into a prompt

The model makes the sound from the same prompt as the picture, so describe what we hear as well as what we see.

  • Speech, on all four Seedance models, FLUX 3 Video and MiniMax H3 (Hailuo 3.0): say who speaks, then put their exact words in quotation marks. Keep it to a line or two for a short clip.
  • Effects: name the sounds that come from the action, such as a door slam, footsteps on gravel or a glass set down.
  • The room or the place: wind, a busy café, rain on a car roof. It tells the model what the air sounds like between the effects.
  • Music, if you want it: the style and the mood, such as a slow piano or a tense drum. Say “no music” if you will add your own.
  • On all four Seedance models, the Sound setting has a “No music” choice. It asks the model to leave the music out; it can't force it, because the model makes all the sound as one track.

Make a clip with sound in Root

  1. Pick a model that makes sound. Add a Video Gen node and choose a model from the table above. Its Sound setting, or a line under the model, says what the clip will sound like. MiniMax H3 (Hailuo 3.0) shows neither.
  2. Set the Sound setting. Choose “On” for a soundtrack or “Off” for a silent clip. On Seedance you can also choose “No music” to keep dialogue and effects only. Veo 3.1 and MiniMax H3 (Hailuo 3.0) always make sound, so there is no setting.
  3. Describe the sound and run it. Add the speech, effects, place and music to the prompt. The price shows on the node before you run it, with sound included.

Related guides

AI video with sound questions

Is there an AI video generator with sound?

Yes, several. On Root, all four Seedance models, Kling 3.0, Kling 3.0 Omni, Kling 2.6, Veo 3.1, FLUX 3 Video and MiniMax H3 (Hailuo 3.0) make sound with the picture. Veo 3.1 and MiniMax H3 (Hailuo 3.0) always do; on the others you can turn it off. The table above shows what each one costs with sound and without.

Does sound make an AI video cost more?

It depends on the model. On all four Seedance models and FLUX 3 Video, no: a clip costs the same with sound. On Kling 3.0 and Kling 3.0 Omni, yes: $0.73 with sound against $0.49 without, for a 5 s, 720p Kling 3.0 clip. On Veo 3.1 and MiniMax H3 (Hailuo 3.0), sound is part of every clip and its price.

Can an AI video model make people talk?

Yes. All four Seedance models, FLUX 3 Video and MiniMax H3 (Hailuo 3.0) make dialogue as part of the clip's sound. Write who speaks and put the words in quotation marks. To put your own recording on a face instead, use lip sync, which runs in the Root app.

Does a clip keep its sound when I cut it into a video in Root?

Only if you add it. Root's timeline export plays only the audio track. Add each take to the picture track with its Video button and to the audio track with its Audio button, which adds the take's whole sound after the last sound already there. Then trim and move each sound so it lines up with its picture.

Can I give the model a sound to work from?

Yes, on Seedance. Seedance 2.5 takes up to 10 audio tracks as references, alongside images and clips, and Seedance 2.0, Seedance 2.0 Fast and Seedance 2.0 Mini up to 3. Name each one in the prompt, for example “@Audio1”, to say how to use it.

How do I add sound to a silent AI clip?

Kling 2.5 Turbo, Runway Gen-4.5 and Luma Ray 3.2 make silent clips. Make a music track in Root, $0.16 whatever its length up to 3 minutes, and put it under the clip on the timeline, or make the shot again with a model that makes sound.

Can I try AI video with sound for free?

For a month, yes. The first month of Root's Beginner plan is free and comes with 650 credits, enough for ten 5 s, 720p Seedance 2.0 Mini clips with sound. After that the plan is $9 a month, and you can cancel whenever you like.

Try Root free for a month

The first month of the Beginner plan is free and comes with 650 credits to spend on any model. After that it's $9 a month, and you can cancel whenever you like.