Your clip is silent by design: the generator makes video only, never music or AI vocals. To add sound, upload the MP4 to TikTok, tap Add sound, search for "Raindance" by Dave featuring Tems, scrub to the bridge where Tems repeats "I love you", trim the sound to your clip, and set your clip's own volume to zero. Reels and CapCut work the same way.
Why does my Raindance AI video have no sound?
Two reasons, and both are deliberate.
First, the AI model generates pictures and mouth movement, not audio. What you get from the generator is an MP4 with a soft singing performance and no audio track, so there is nothing to mute or balance.
Second, pairing a song with a video outside the platform is the fastest way to get muted. TikTok and Instagram license music for use inside their own apps, so a sound attached from the in-app library is cleared — and it carries the sound name, which makes your post discoverable by people tapping through the trend. An MP3 baked into the file before upload gets flagged by the audio-fingerprint check instead.
So: render the video silent, add the sound inside the app. That order is the whole trick.
How do I add the Raindance sound on TikTok?
- Export your clip and save it to your camera roll. A 9:16 vertical file is what you want; TikTok letterboxes anything else.
- Open TikTok, tap +, then Upload, and pick your clip.
- On the edit screen, tap Add sound at the top (it reads Change sound if one is already attached).
- Search Raindance. The official track is credited to Dave featuring Tems. AI covers and sped-up edits share the name, so check the artist line and cover art — the use counts under a sound tell you which version the trend is on.
- Tap the sound, then the scissors or trim icon, and drag the waveform to the bridge where Tems repeats "I love you". It sits well past the first minute, so scrub rather than accepting the default start point.
- Trim the sound to your clip's length. An 8-second video needs about 8 seconds of audio.
- Tap Volume (sometimes under Sound or Edit sound). Push Added sound to around 100% and Original sound to 0%. Your clip is silent anyway, but zeroing it avoids hiss from a re-encode.
- Tap Next, write your description, then post.
If the start point keeps snapping back, move the video instead: shift your clip later in the timeline so the bridge lands on the push-in. The strongest timing on an 8-second clip is the hard cut to the pier arriving just before the vocal does.
How do I add it in Instagram Reels?
- Save the MP4 to your camera roll.
- In Instagram, tap +, choose Reel, and select the clip.
- Tap the music note (Add audio), search Raindance, and pick the Dave and Tems original.
- Tap the waveform to open the audio trimmer and drag to the "I love you" bridge.
- Open the mixer and set your original audio to 0%.
- Share. Reels also lets you pick a cover frame — the first frame of the pier shot usually reads better than the opening recreation of your photo.
Instagram's catalogue varies by region and account type. If the full track is missing on a business account, switching to a creator account usually restores it.
How do I add the sound in CapCut?
Use CapCut when you want control the social apps do not give you: fades, a beat-matched cut, or a 15-second edit.
- Open CapCut, tap New project, and import your clip.
- Tap Audio, then Sounds, and search Raindance. CapCut's own library is licensed for editing, but tracks there are not always cleared for posting elsewhere.
- Trim to the bridge, then drag the audio clip so the vocal lands on the cut.
- Tap the clip, then Volume, and set it to 0.
- Export at 1080p, 30fps.
- Here is the catch: a CapCut export with music baked in can still be muted on TikTok. The safest route is to export your edit from CapCut without audio, then add the sound in TikTok as above. Use CapCut for the cutting, TikTok for the music.
CapCut also hosts ready-made trend templates — the two biggest are named "i love you" and "Raindance", with roughly 271.9K and 179.9K uses at the time of research (CapCut's Rain Dance page). They slot your photos into a pre-cut edit; they do not generate an AI person singing. See the CapCut template guide for the difference.
| Where the music comes from | Best for | Mute risk | |
|---|---|---|---|
| TikTok | In-app library, attached at post time | Posting to TikTok | Lowest |
| Instagram Reels | In-app audio library | Posting to Reels | Low |
| CapCut | Editor library, baked into the export | Timing, fades, longer edits | Higher if you post the baked file |
TikTok muted my Raindance video — how do I fix it?
A muted post almost always means the audio was embedded in the file rather than attached in the app. Fix it like this:
- Delete the post. You cannot swap the sound on a live TikTok, and editing the caption will not restore audio.
- Re-upload the silent MP4 and add the sound from the library, following the steps above.
- Before you publish, run TikTok's music copyright check on the post screen. It flags audio that will be removed, so you can fix it while it is still a draft.
- If you believe the mute was wrong — for example you used a library sound and it was pulled anyway — you can appeal from the video's Manage or notification screen. TikTok documents both the removal reasons and the appeal flow in its sound-removal help article and the muted-video appeal guide.
Two smaller causes: a regional catalogue that lacks the track, and a business account, which only gets the commercial-use library. Switching to a creator account restores the full catalogue.
Do I have to label my Raindance video as AI on TikTok?
Yes. TikTok requires realistic AI-generated content to be labelled, and a video of you singing on a pier that you never filmed is exactly that. Turn the label on before posting: on the final post screen, open More options and switch on the AI-generated content toggle. Some uploads are detected and labelled automatically, but do not rely on it — an unlabelled realistic AI video can be removed.
Labelling costs you nothing in reach. Instagram does the same thing with its own AI-info label.
Two related rules: use your own face, or the face of an adult who has agreed to it — never a celebrity, a stranger or a child — and if you are making a two-person version, ask first. The duet guide covers that.
What if the sound still will not line up?
The model cannot hear the song, so the mouth movement is a soft performance rather than frame-accurate lip-sync. Changing the audio will not fix it, but you can hide it: start the sound on the beat before the cut, keep the clip short, and pick a take where the singing is gentle rather than wide-mouthed. The photo guide covers which source photos perform best, and the free teaser lets you check the timing before you spend anything. Clips and bundles are on pricing.