pipeline-examples/talking-avatar 🖼️✓❓ → 🖼️

▶️ 239 runs 📅 Aug 2025 ⚙️ Cog 0.14.0

About

A lip-syncing avatar for UGC/TikTok/Reels content

Example Output

Output

Performance Metrics

401.07s Prediction Time

401.58s Total Time

All Input Parameters

{
  "audio": "https://replicate.delivery/pbxt/NWYEr8QPkBeQZaJ2GkPlU1z1B7CWQJtyUJRoUnpwlluZaZ6g/replicate-prediction-gz0dq0zj71rma0crky99mj4q0r.mp3",
  "image": "https://replicate.delivery/pbxt/NZOWZquQIRTicbBpH4PAUBAasf0ca0tVdp25TF7riFBUayUw/GyqIvuYW8AsNRiW.jpg",
  "captions": true,
  "duration": 10,
  "resolution": "720p"
}

Input Parameters

audio (required) Type: string: Audio file for lip synchronization (.mp3, .wav, .m4a, or .aac)
image (required) Type: string: Input image of a person for the talking avatar
captions Type: booleanDefault: true: Return video with captions
duration Default: 10: duration of the video in seconds
resolution Default: 720p: Video resolution

Output Schema

Output

Type: string • Format: uri

Example Execution Logs

Generating video from image...
Adding lip synchronization...
Adding captions...

Version Details

Version ID: b08b9a00beb1cc4df32751294d98b3147604fe92e6a16d074683f462e5efd838
Version Created: August 20, 2025

Run on Replicate →