VIDOME — lyric videos, timed to your vocalThe mix is
done. The
video isn’t.
Upload the track. Paste the words. Vidome listens to the vocal and lands each word on the beat it is sung on — then you nudge the two it got wrong, choose a look, and export it wide for YouTube and tall for everything else.
Built for the fortnight before a release. Vidome is a web app that turns a song and its lyrics into a finished lyric video: upload a track, paste the words or let it transcribe them, and Vidome syncs each word to the vocal, styles the result, and exports it in landscape or portrait for YouTube, TikTok and Reels.
20 credits on signup, no card · what it costs after that
Playing now
The shipping renderer, in this tab · 13-second loop · no audio
0:00.00 / 0:13.00
SpotlightProScrolling stack with a soft bloom on each word. The streaming-app look.
Word timing, live
between lines
- I0:00.40
- remember0:00.70
- when0:01.22
- the0:01.45
- lights0:01.64
- went0:02.04
- down0:02.30
Why bother
A lyric video is
not a music video.
Nobody is asking you to direct anything.
It is the version of the song that works with the sound off. It is what turns up when somebody half-remembers a line and types it into a search box. It is the thing you cut a fifteen-second vertical out of at seven in the morning on release day. Least glamorous asset on the whole campaign, most played.
And what it normally costs you is a weekend. Not a good weekend — a weekend spent dragging keyframes until somewhere sits under somewhere, then finding out at bar sixty that the second verse has been a beat and a half late the entire time.
We built the boring half. You keep the half that is a decision.
How the timing is worked out
Speech models mishear singing. Ours does too — which is why it never gets to choose the words. Only the moment.
You paste the lyrics, the model returns timed guesses, and an alignment pass keeps nothing from it but the clock — so a mangled transcript can cost you timing, and can never change what is on screen.
How it works
The whole job, start to upload
Four moves, and none of them a wizard you cannot back out of. Change the lyrics at move four and the timings follow them.
Drop the track in
MP3, WAV, M4A, FLAC or OGG
The waveform is read on the way in, in buckets of about 50 milliseconds, so scrubbing is instant afterwards and the reactive backgrounds move to your low end rather than to a timer. Listening to the vocal is charged once per file — upload the same song again, or have a bandmate upload it, and it is free: files are matched by checksum, not by name.
Give it the words
Paste them, or let it listen
Paste them if you have them — your spelling, your capitals, your line breaks, your call on “gonna” versus “going to”. [Chorus] and (Verse 2) are read as structure and never appear on screen. If you do not have them written down, leave the box empty and it will write them down off the vocal for you to correct.
Pick a look, then argue with it
A preset is a starting point
A theme sets the face, the ground and the motion, and then every one of those is yours to change: weight, tracking, colour, stroke, glow, the plate behind the line, how a word arrives, how it lights up as it is sung, how it leaves. Or drop in your own artwork or a video clip and the lyrics composite over it — a short clip loops to cover the track.
Export both shapes
16:9 and 9:16, one project
Wide for YouTube, tall for Reels, Shorts and TikTok. The type rescales and re-wraps for the frame, so the vertical cut is a second export rather than a second edit.
Word by word
Not line by line.
Word by word.
Every word carries its own in and its own out — the difference between a wipe that tracks the vocal and a caption that changes somewhere near the right bar.
between words
2 of these words start out marked amber, and they are staged: the line was written by hand, so nothing in it was genuinely estimated. In the app that mark means the aligner could not match a word and worked its position out from its neighbours.
This is the editor’s own timeline, running the line the player above is singing. Press play, then click a word at the moment you would have heard it — that is tap-to-sync, and it is the fastest fix there is, because your ear does the job instead of your eye. Or shift the whole line by 50 milliseconds at a time when everything is running a shade behind.
Highlighting lands 55 milliseconds ahead of the vocal. Sat at dead zero it reads late — the eye wants a moment the ear does not.
Where you start
26 places to start,
none of them a lock.
11 of them are on the free plan, and pressing any row plays it in the frame at the top of the page.
Applying a theme writes its values into your project and then gets out of the way — there is no preset left underneath to fight with. Beneath all of them sit 13 ways for a word to arrive, 14 ways for it to light up as it is sung, and 14 typefaces.
What it costs
Three minutes,
15 credits.
3 to listen to the track once, 12 to render it at 1080p. You are given 20 for making an account.
One export, priced · 3-minute track
- Preview render
- 0
- Watermarked, and never rationed. This is how you decide.
- 720p · 30 fps
- 8
- The ceiling on Free, and where the watermark stays.
- 1080p · 30 fps
- 12
- Clean export from Starter up.
- 1080p · 60 fps
- 23
- Twice the frames, so nearly twice the cost.
- 4K · 30 fps
- 39
- Pro and above.
A video background adds a decode pass, so it is 1.25× the figures above. The export dialog budgets about 22 seconds of machine time for that 1080p render, plus whatever the queue is doing ahead of you. If a render fails you are refunded without asking.
The free plan, with nothing left out
- Credits
- 20once, not monthly
- Export ceiling
- 720p30 fps, watermarked
- Track length
- 4 mincovers most singles
- Upload size
- 25 MBper file
Not 20 a month — once. That is one song listened to (3) and exported at 720p (8), then exported again after you have changed your mind (8). Enough to answer the only question that matters: does the timing hold up on your vocal, at your tempo.
Where it struggles
Four things we would rather you heard from us.
- A loud mix is where it does worst
- Doubled vocals, heavy reverb and a wall of guitars all hurt the transcript, and the transcript is what carries the timing. The line still lands; individual words inside it are the ones you end up tapping. A clean vocal up front and you will rarely touch it.
- Every free export carries the watermark
- Free tops out at 720p and the mark stays on. Preview renders are watermarked for everyone, on every plan, always — that is what makes them free. We would rather you read that here than find it after your first export.
- The free credits do not come back
- 20 credits once, when you sign up, not 20 a month. A monthly allowance on a paid plan expires at the end of the period; credits you buy in a pack never do.
- One tool, so far
- Vidome makes lyric videos. There are 4 more tools on the roadmap and not one of them is finished, so do not sign up for those yet.
Pricing
Two things
cost credits.
Listening to a track, once. And exporting a finished video. Editing, restyling and watermarked previews are free and always will be, so you only pay for a render you have already watched.
| Feature | Free$0 forever Start free | Starter$9 per month Choose Starter | First with teams, 4K, 60 fps Pro$29 per month Choose Pro | Studio$99 per month Choose Studio |
|---|---|---|---|---|
| Credits | 20 one time | 60 / month | 250 / month | 1,000 / month |
| Max export quality | 720p | 1080p | 4K | 4K |
| Frame rate | 30 fps | 30 fps | 60 fps | 60 fps |
| Track length | 4 min | 10 min | 15 min | 30 min |
| Upload size | 25 MB | 100 MB | 250 MB | 1000 MB |
| Storage | 1 GB | 10 GB | 50 GB | 500 GB |
| Renders at once | 1 | 2 | 4 | 12 |
| Exports without a watermark | Not included | Included | Included | Included |
| Pro effects, themes & fonts | Not included | Included | Included | Included |
| Custom brand watermark | Not included | Not included | Included | Included |
| Team seats | Not included | Not included | 3 included | 10 included |
| API access | Not included | Not included | Included | Included |
| Priority rendering | Not included | Not included | Included | Included |
| Priority support | Not included | Not included | Not included | Included |
Top-ups
Need more? Top up any time.
Purchased credits never expire, and your monthly allowance is always spent first. A paid allowance does expire at the end of each period.
250 credits
$4.80 per 100$12
1,000 credits
$3.90 per 100$39
A 3-minute song is 15 credits all in: 3 to listen to it once, then 12 to render it at 1080p. Preview renders cost nothing, and a failed render is refunded without asking.
Last thing
Put the words
on the screen.
20 credits when you sign up, and no card. That is one song listened to and exported at 720p, twice over — enough to find out whether the timing holds on your own vocal.
Make the first oneThen come back and tell us which words we got wrong.
You can create a Vidome account with an email address and password, or by signing in with Google. If you choose Google, Vidome receives your name, email address and profile picture, and uses them for one thing: to create your account and identify you when you come back. Vidome does not read, send or store anything else from your Google account, and it never posts anywhere on your behalf.
Your songs, lyrics and finished videos belong to you. What we keep, how long we keep it, and how to delete all of it is set out in the privacy policy.