Lip Sync (Video)

Sync an existing video’s lip movements to new audio — perfect for dubbing, dialogue replacement, or creative remixes.

Lip Sync (Video)

Replace the audio of a video and match the lips.
This node takes existing video footage and re-syncs the mouth movements to new audio — perfect for dubbing, dialogue replacement, or creative storytelling.

---

🪄 When to Use It

Use Lip Sync when you have video footage and need to change what’s being said.

It works great for:

• Dubbing — translate dialogue to other languages
• Dialogue replacement — change what a character says
• Creative remixes — make characters say new things
• Voice corrections — fix audio without reshooting

Skill level: Advanced
Average time: 90–180 seconds depending on length
Cost: High

---

🎚️ Controls and Parameters

Video (Video, Required)

The footage you want to re-sync. Works best with:

• Clear face visibility throughout
• Minimal camera movement
• Good lighting on the face

💡 Tip: Close-up or medium shots work better than wide angles.

Audios (Array, Required)

New audio to sync to the video. Supports up to 2 speakers:

• Audio 1 — main speaker (left side of frame)
• Audio 2 — second speaker (right side of frame)

💡 Tip: If you only have one speaker, just connect Audio 1.

Prompt (Text, Optional)

Describe the context: “Two people having a conversation” or “Single person speaking to camera”

Audio Mode (Select)

• Sequential — speakers take turns (dialogue back and forth)
• Parallel — both audio tracks play simultaneously, allowing precise timing control

💡 Tip: Use Sequential for back-and-forth dialogue. Use Parallel when you need exact timing control — each audio track can have silence during non-speaking parts.

Example (Parallel mode):

• Speaker 1: 10-second audio with dialogue for first 5 seconds, then 5 seconds of silence
• Speaker 2: 10-second audio with 5 seconds of silence, then dialogue for last 5 seconds
• Result: Speaker 1 talks, then Speaker 2 responds, with perfect timing control

Seed (Number)

Control randomness for consistent results.

---

🎨 Available Models

Choose the model that fits your project needs:

Infinite Talk (Default)

Advanced multi-speaker lip-sync with support for up to 2 simultaneous speakers. Handles complex dialogue and maintains facial expressions.

Features:

• Sequential or parallel audio modes
• Custom prompts for scene context
• Seed control for consistency

PixVerse

Alternative lip-sync engine optimized for speed. Good for quick turnaround projects.

💡 Tip: Use Infinite Talk for professional work with multiple speakers. Try PixVerse for faster processing.

---

🎨 What to Expect

Lip-sync AI will:

• Retime mouth movements to match new audio
• Attempt to preserve facial expressions
• Keep the rest of the video unchanged

Best results with:

• Clear frontal face shots — direct view of mouth
• Consistent lighting — no dramatic shadows on face
• Audio matches video length — similar duration helps quality

Challenges with:

• Profile views or turned heads
• Very fast dialogue with lots of mouth movement
• Poor video quality or compression artifacts
• Multiple people talking when only one audio is provided

---

💬 Quick Tips

• Audio length should roughly match video length for best sync
• Use Voice Generator to create matching dialogue audio
• Test with short clips first before processing long videos
• If results look off, try adjusting the prompt to describe the scene better
• Works best with videos that have minimal head movement