Lip Sync Intermediate

Runway's Lip Sync tool modifies the mouth movements in existing video to match a new audio track. This enables dubbing AI-generated characters, re-voicing existing footage, or adding dialogue to silent video — all with convincing, natural-looking lip movements.

How It Works

Upload a video
Provide a video clip with a visible face. This can be a Runway-generated clip or any uploaded footage.
Provide audio
Upload an audio file (WAV, MP3) with the speech you want the face to match.
Generate
Runway analyzes the audio phonemes and modifies the face in the video to produce matching mouth shapes, jaw movement, and subtle facial expressions.

Input Requirements

Input	Requirement	Tip
Video	Clear face, front-facing preferred	Minimize head rotation during speech
Audio	Clean speech, minimal background noise	Use noise reduction before uploading
Duration	Audio and video should be similar length	Trim to match before processing
Resolution	720p minimum	Higher resolution = better lip detail

Use Cases

AI Character Voiceover: Generate a character with Gen-3, then add dialogue with Lip Sync
Language Dubbing: Re-voice existing content in a new language with matching lip movements
Content Correction: Fix or update spoken content without re-shooting
Creative Projects: Make photos or illustrations "speak" by first animating with Gen-3 then applying Lip Sync

Multi-Step Workflow: For the best results with AI-generated characters, use a two-step approach: first generate the character video with Gen-3 Alpha (focusing on visual quality), then apply Lip Sync as a separate pass (focusing on audio accuracy). This gives you independent control over visual and audio quality.

Limitations

Lip Sync works best with front-facing, well-lit faces speaking at moderate pace. Extreme head angles, fast speech, or heavily occluded faces may produce artifacts. Always preview the output and regenerate if quality is insufficient.

← Motion Brush Workflows →