We’re excited to introduce the VEED Lipsync API: the world’s most powerful lip-syncing technology, now available for developers and teams. The API takes any video and remaps the speaker’s lip movements to match a new audio track. Submit a video URL and an audio URL, and it returns a synchronized MP4. You can access it today on fal.ai.
Key takeaways:
- The VEED lip sync API is a video-to-video API that replaces lip movements to match new audio
- It accepts MP4, MOV, WebM, M4V, and GIF video files plus MP3, OGG, WAV, M4A, and AAC audio files
- Two inputs are required: a video URL and an audio URL
- The API returns an MP4 with synchronized lip movements as a JSON response
- Pricing is $0.40 per minute of processed video, accessed through fal.ai
What does our Lipsync API do?
The VEED lip sync API analyzes facial movements in a source video and remaps lip positions to match a replacement audio track. It handles natural mouth shapes, speech timing, and facial dynamics automatically. No manual animation or frame-by-frame configuration is required.
We built this for teams that produce video at scale: dubbing pipelines, localization workflows, AI avatar systems, and content rephrasing. The API processes videos at approximately 2 to 2.5 minutes of generation time per minute of video input.
Here's a sample AI video showing our tech in action:
API documentation
The VEED lip sync API is hosted on fal.ai under the veed/lipsync endpoint. Below is everything a developer needs to evaluate and integrate the API.
Authentication
The API uses fal.ai's authentication system. Set your FAL_KEY as an environment variable in your runtime:
export FAL_KEY="your-fal-api-key"
Get your API key by signing up at fal.ai.
Endpoint
fal-ai endpoint: veed/lipsync
Full model path: https://fal.ai/models/veed/lipsync
Input parameters
Two parameters are required:
- video_url (required): URL of the source video. Accepted formats: MP4, MOV, WebM, M4V, GIF.
- audio_url (required): URL of the replacement audio. Accepted formats: MP3, OGG, WAV, M4A, AAC.
The API accepts any video aspect ratio. If your audio and video durations are out of sync, fal.ai's lipsync mode defaults to cut_off (trims to the shorter of the two). Other sync modes include loop, bounce, silence, and remap.
Sample request (JavaScript)
Install the fal client:
npm install @fal-ai/client
Submit a request:
import { fal } from "@fal-ai/client";const result = await fal.queue.submit("veed/lipsync", { input: { video_url: "https://your-video-url.mp4", audio_url: "https://your-audio-url.mp3" }});
Response format
The API returns a JSON object. The output video is accessible at response.video.url:
{ "video": { "url": "https://v3.fal.media/files/output.mp4", "file_name": "output.mp4", "file_size": 1593113, "content_type": "application/octet-stream" }}
Pricing
The VEED lip sync API costs $0.40 per minute of processed video. A 5-minute video costs $2.00. Pricing is usage-based with no seat fees or monthly minimums. For enterprise volume discounts, contact VEED's sales team.
Key use cases
- Dubbing and localization: Create localized videos that feel natural across languages.
- Video rephrasing: Swap out words or sentences in pre-recorded content without re-shooting.
- AI avatars: Generate avatars that can speak dynamically in sync with new audio.
Get started with the VEED lip sync API
The lip sync API is available now on fal.ai. Sign up for a fal.ai account, set your FAL_KEY, and you can run your first request in minutes. Additional provider integrations are coming soon.
For enterprise use, contact our sales team to request a demo and discuss volume pricing. You can also explore VEED's full AI video creation platform for browser-based editing tools alongside the API.
.png)