One-Click AI MV Generator
Upload a track, choose a look, and get a lip-synced music video with burned-in subtitles. Credits are billed per second, so you only pay for the length you actually use.
Nothing generated yet
Your MV appears here as soon as the first frame is ready.What is a one-click AI MV generator?
A one-click AI MV generator turns an audio track into a finished music video. Upload a song, give the model a performer, and it produces footage cut to the music with the vocals lip-synced frame by frame — no timeline, no keyframes, no editing software.
OpenFilmo runs this on top of Vidu's one-click MV models. Two generations are available: V1 for general-purpose music videos at up to 1080p, and V2 Premium for performance-driven clips that alternate between singing and acting. Both accept a style prompt and can burn subtitles derived from the lyrics.
How to make a music video in three steps
- Upload your track
Drop in an mp3, wav, m4a or aac file and drag the two handles to pick the {min} to {max} second section you want to use.
- Choose the performer
Add up to seven character images, or two face references if you are using a V2 singing mode. Order the faces the way they appear in the song.
- Generate and download
Pick a resolution and aspect ratio, then generate. The clip usually renders in one to three minutes and downloads as an MP4.
V1 vs V2 Premium
| Feature | V1 | V2 Premium |
|---|---|---|
| Resolutions | 540p, 720p, 1080p | 720p, 1080p |
| Modes | Single general mode | Sing, sing + perform, perform, dance |
| Lip sync | Optional add-on (+4 credits/second) | Built into every mode |
| Best for | Simple lyric videos and quick experiments | Multi-shot performance videos with acting beats |
How MV credits are calculated
MV credits are linear in duration: credits = per-second price × seconds. The per-second price depends on the version, the V2 mode and the resolution, and lip sync on V1 adds 4 credits per second. Because the audio decides the duration, the server re-measures the uploaded file rather than trusting what the browser reports. Off-peak discounts do not apply to MV.
Frequently asked questions
- How long can the music be?
- Between {min} and {max} seconds per generation. Longer tracks are trimmed to {max} seconds at submit time, so pick the section you want first.
- Which audio formats are supported?
- mp3, wav, m4a and aac. The format is detected from the file header, so a mislabelled extension will still work as long as the container is one of those four.
- Do I need to upload a face for lip sync?
- For V1 lip sync, no — the model works from the character images. For V2 singing modes, yes: up to two face references, ordered by who sings first.
- What happens if my credits run out mid-task?
- Credits are charged once when the task is created, not during rendering. If the generation fails on the provider side, the charge is refunded automatically to the same account.
- Can I use the result commercially?
- You own what you generate, but you are responsible for the rights to the audio and any images you upload. Do not upload music you are not licensed to use.
- Why does the page ask me to re-select the clip length?
- The server re-reads the uploaded audio to compute the billed duration, because that number decides the price. If your browser and the server disagree, the server value wins.