Dawnwood / Workshop / Auto Sound
SOUND, PACKAGE AND POST · FOR STITCHED AI VIDEO FROM SEEDANCE, KLING, VEO

Sound it, package it, post it.

Every clip from Seedance, Kling or Veo comes with its own music, so a stitched video stops and restarts its music at every cut. Auto Sound fixes the sound and gets the video posted, all on one page:

  1. Sound. It strips the music baked into each clip, keeps the dialogue and scores the whole video once, with sound effects across every cut.
  2. Package. Add pop captions, pick a cover from four frames and start from a drafted post with hashtags. Included in the price.
  3. Post. Send it to your own TikTok, Instagram or Facebook, now or at a suggested time. Your first 3 videos post publicly with no card.
Long videos and full episodes supported · first 3 videos free (up to 1 minute each) · then 5 cr per minute of video · captions, cover and posting included · failed jobs refunded
AUTO SOUND, PLAINLY 1 score across every cut Dialogue kept Long videos full episodes Post to TikTok, Instagram, Facebook 5 cr per minute
HEAR IT

Seven clips, one score.

Seven Seedance clips from our drama The Bought Bride, cut into one scene. Before: the clips as rendered, so the sound comes and goes at every cut. After: one pass of Auto Sound with no hand edits. A standoff, then battle music the moment her power breaks out, then a standoff again when he steps in. Turn your sound on and switch while it plays.

Added automatically: one score in 3 sections · 12 effects · ballroom ambience

Try it on your video, free

THE ANSWER

Why the music breaks, and what Auto Sound does.

Why does the music cut out between AI video clips? AI video models such as Seedance, Kling and Veo make clips of 30 seconds or less and score each one on its own. Every clip carries music that starts and ends with it, so when you stitch clips into a scene the music stops and restarts at every cut, and the mood jumps from tense to romantic to epic in twenty seconds.

What is Dawnwood Auto Sound? Auto Sound removes the background music from each clip, only where music is actually heard, and keeps the dialogue and the rest of the track. Then it watches the whole timeline at once and scores it as one piece: the music runs straight across your cuts and changes only where the story turns, from a standoff to a fight or from a fight to a rescue. It ducks under every line, adds the sound effects the scene is missing and lays an ambience bed under each clip, without doubling a sound that is already there.

It was built for short drama made from many short AI clips, and it works the same on any video you upload. Want to do it by hand? Our guide why the music cuts out between AI video clips walks through the manual fix with free tools.

THE FIX

Strip, keep, score.

01 · STRIP
Remove each clip’s music
Clips where music is heard are separated and their music removed. Clips with no music are left as they are.
02 · KEEP
Dialogue stays untouched
Voices, lip sync and the sounds the model already made stay on the track, so nothing has to be re-voiced.
03 · SCORE
One score for the whole video
The score runs across every cut and changes only at story turns, with sound effects and ambience laid in and the music ducked under the lines.
PACKAGE AND POST

From finished sound to a live post.

01 · CAPTIONS
Pop, plain or none
Word-by-word pop captions are on when the video has dialogue and off when it has none. The frame stays as you made it unless you choose Fit to 9:16.
02 · COVER AND TEXT
Ready before you touch it
Four suggested cover frames and a one-line post with hashtags, drafted from the dialogue and what happens on screen. Change anything, or skip and post the video as is.
03 · POST
Your own accounts
Connect TikTok, Instagram or Facebook without leaving the page and post now or at a suggested time. You see each post go live, and we email you the views a day later. YouTube: download the video and upload it yourself.

Captions, the cover, the post text and posting cost nothing beyond the Auto Sound price, and changing the captions re-renders for free. Your first 3 videos post publicly with no card, one a day; one video sent to several accounts counts once. After that, a $0 card check (no charge) lets you keep posting. A download of the packaged video is always one click away.

LONG VIDEOS

Full episodes, not just short clips.

Many music tools stop at a few minutes of video. Auto Sound is made for full episodes stitched from dozens of clips. Each pass handles up to 15 minutes of footage it has not seen before; add the next part to the timeline and run it again. Footage already done is not charged twice, and the score is laid across the whole timeline, so the same mood keeps the same music from the first scene to the last.

HOW IT WORKS

Upload, listen, fix, post.

01 · SOURCE
Upload or use your timeline
Upload a video on the Auto Sound card in Tools, or press Auto sound in the editor to work on the clips already on your timeline.
02 · LISTEN
Picture and track together
It finds what the picture shows that the audio lacks: impacts, doors, glass, footsteps, weapons, creatures, magic, and sounds from just off screen.
03 · FIX
Every effect is editable
Each effect lands on the Sound track with its reason. Swap it for one of three alternatives, change the gain, nudge a frame, lock it or reject it.
04 · POST
One finished video, out the door
The mix plays as soon as it is ready. Finish & post adds captions and a cover and sends it to your accounts, or download it. Effects follow the footage when you trim, move, split or duplicate clips in the editor.
SUBTLE

Fewer effects, only the ones it is most sure about. For dialogue-heavy scenes and quiet drama.

STANDARD

The default. Every clear action in the picture gets its sound, nothing more.

PUNCHY

More effects at a lower confidence bar. For fights, transformations and anything that should hit hard.

THE DETAILS

What a good sound pass needs.

Keeps what is there

The existing track is checked first. If the door already slams, Auto Sound does not add a second slam.

Stays out of the way of speech

Inside dialogue it adds only confident impacts, and it never adds effects over foreign-language or dubbed lines.

Scored as one piece

Music is spotted on the whole timeline, not clip by clip. Each section lasts at least 8 seconds, the same mood keeps the same track, and sections crossfade.

Rejections stick

Mark an effect as don’t want this and it stays out the next time you run the pass. Locked effects are left alone.

Layers you can switch off

Already have a score you like? Turn the music layer off and the clip audio is left as it was. Effects, ambience and music are separate switches.

Agents can run it

Over MCP your agent uploads a video, starts a pass, checks on it and exports the result. See For agents.

IN A DAWNWOOD SERIES

Every new episode gets its sound pass.

On series made in Dawnwood, the same engine runs on every new episode master by itself. Series created from November 9, 2026 have it on; older series can switch it on. The episode page has a Sound panel where you check and fix the effects the same way as in the editor.

IN DISCORD

Use it in Discord.

Add the Dawnwood app to your Discord account, type /sound in any server or DM and attach a video. The finished video with sound is posted back in the channel, usually within a few minutes. It uses the same Dawnwood account, price and free tries as this page. In the Dawnwood Discord server your first 3 runs are free. How it works in Discord.

Add to Discord Join the Discord server
PRICING

Pay once per minute of footage.

5 cr per minute of video
Auto Sound costs 5 cr per minute of video, rounded up per started minute, with a 1 minute minimum. You pay only for footage the project has not been through before, so running it again at another intensity is free. A job that fails is refunded in full. Captions, cover, post text and posting are included. Exporting from the editor is 1 cr. Credits come with every plan, and new accounts start with 500.
VideoCredits
A 45-second clip5
A 1:20 drama episode10
A 12-minute cut60
The same 12 minutes again, set to Punchy0
QUESTIONS

Before you upload.

Why does the music cut out between my AI video clips?
Seedance, Kling and Veo score each clip on its own, so every clip carries music that starts and ends with it. Stitched together, the music stops and restarts at every cut and the mood jumps. Auto Sound removes the music from each clip, keeps the dialogue, and lays one continuous score across the whole video that changes only where the story turns.
What does Auto Sound add, and what does it leave alone?
It listens to the audio first and keeps the sounds that are already there, so nothing is doubled. Then it adds what the picture shows but the track lacks: hits and slaps, doors, glass, footsteps on distinctive surfaces, weapons, creature sounds such as growls and howls, magic or power bursts, and off-screen sounds like a door shaking. Each clip also gets a looping ambience bed, and a music bed from the shared library is ducked under the dialogue.
Will it put effects over my dialogue?
Inside dialogue it only adds impacts it is confident about. It never adds effects over foreign-language or dubbed dialogue windows.
Can I control how much it adds?
Pick Subtle, Standard or Punchy. That sets how many effects it adds and how confident it must be before adding one. You can also switch off the effects, the ambience or the music layer.
What if one effect is wrong?
Every effect sits on the editor's Sound track with the reason it was added. Preview three alternatives and swap, change its gain, nudge it one frame either way, lock it, or mark it as don't want this. Rejected effects stay out on the next run.
What happens to the effects when I edit the video?
They stay attached to the footage. Trim, move, split or duplicate a clip and its effects go with it. Export renders the finished video with the sound mixed in.
Will the effects be audible on a phone?
Effects that sit mostly in low frequencies get a mid and high presence layer, so a punch or a slam still reads on phone speakers.
How long can the video be?
Long videos are supported, so you can run full episodes. Each pass handles up to 15 minutes of footage it has not seen before; add the next part and run it again, and footage already done is not charged twice.
What does it cost?
5 cr per minute of video, rounded up per started minute with a 1 minute minimum. You pay only for footage the project has not been through before, so running it again at a different intensity is free. Failed jobs are refunded in full. Captions, the cover, the post text and posting are included.
Where can I post it?
To your own TikTok, Instagram and Facebook accounts, which you connect from the same page. YouTube isn't open for posting from Dawnwood yet, so download the packaged video and upload it yourself. Your first 3 videos post publicly with no card, one a day; after that, add a card ($0 check, no charge) to keep posting.

Music that cuts out makes cheap scenes.

Upload your stitched clips, hear one score run across every cut, and post the finished video to your own accounts. You pay by the minute of footage, and a failed job costs nothing.