Reference to Video with Lip Sync
Use reference images to lock the look of your characters and scenes, and your own audio as the soundtrack — lip movements are aligned to the first audio you upload.
API key
Enter your API key to submit or inspect tasks.
Create a task
Fields map directly onto POST /aigc/task with task_type: "ref2va_lipsync".
Lock the look of people, scenes and objects. PNG / JPEG / WebP.
None added.
Optional motion / camera reference; picture only, its sound is ignored.
None added.
Up to 2. The first audio becomes the video's soundtrack and the lips follow it; the second is only a tone / style reference.
Up to 9 assets in total. Uploads are limited to 30MB each — use a URL for larger files.
Estimate from the current price table (see the Billing tab). Account-specific discounts may lower the actual charge.
Or inspect an existing task
Request log
Every call this page makes to the API, in order, most recent first.
Nothing sent yet.
How billing works
- Billed per second of generated video, by resolution tier. The full amount is deducted the moment you submit the task — there is no separate reservation or settlement step.
- If the task fails or times out, the full amount is refunded automatically.
- ref2va and ref2va_lipsync share the same model and the same price. Your first reference audio is used as the soundtrack and the lips follow it.
- Aspect ratio, number of reference assets, sampling steps and audio start do not change the price. Only resolution and duration do.
- Invalid requests are rejected with HTTP 400 and are never charged. Task queries are free.
- Accounts with a negotiated rate are charged that rate instead of the list price below.
Price per second
| Resolution | Price per second | Shortest task (3s) |
|---|---|---|
| 540p | $0.015 | $0.045 |
| 720p | $0.03 | $0.09 |
Resolution is nominal: the actual output is slightly smaller (see the sizes table below) and is billed at the nominal tier.
Formula
price = unit_price[resolution] x duration
Duration is an integer number of seconds between 3 and 20. Results are rounded to 4 decimal places.
Worked examples
| Resolution | Duration | Calculation | Price |
|---|---|---|---|
| 540p | 5s | $0.015 x 5 | $0.075 |
| 720p | 5s | $0.03 x 5 | $0.15 |
| 720p | 10s | $0.03 x 10 | $0.3 |
| 720p | 20s | $0.03 x 20 | $0.6 |
Output sizes
| aspect_ratio | 540p | 720p |
|---|---|---|
| 16:9 | 832×512 | 1216×704 |
| 9:16 | 512×832 | 704×1216 |
| 1:1 | 640×640 | 896×896 |
| 4:3 | 768×576 | 1088×768 |
| 3:4 | 576×768 | 768×1088 |
| 3:2 | 768×512 | 1152×768 |
| 2:3 | 512×768 | 768×1152 |
Videos are 24 fps H.264 MP4 with an audio track.