Prompt
Aspect Ratio
Reference Image optional
🖼
Drop or click to add a reference image
Optional — guides the style of the output
Model
✦ Draft (fast & cheap)
★ Production (client & portfolio quality)
More Models
Output Size
1024 × 1024
Small
Max
Prompt
Negative prompt ▾
Style Reference
🎨
Drop style reference image here
or click to browse
Character Reference (up to 3)
👤
Drop character reference images (up to 3)
or click to browse · multiple allowed
Art Style
Batch Count
🎨 Style Reference A (sets look & feel)
🎨
Image A — Style Reference
Color palette, rendering style, art direction
📏 Composition Reference B (preserves structure)
📏
Image B — Composition Reference
Pose, framing, layout to preserve
Model
✦ Draft (fast & cheap)
★ Production (client & portfolio quality)
Aspect Ratio
Output Size
1024 × 1024
Small
Max
Output
🖼
Your generated image will appear here
History
Animation Mode
📝
Text → Video
Prompt-driven generation
🖼
Image → Video
Animate your frames
FIRST FRAME
🟢
Drop first frame
LAST FRAME
🔴
Drop last frame
Prompt
Character Reference (up to 2)
👤
Drop character ref images (up to 2)
Style Reference
🎨
Drop style reference (1 image)
Duration
Camera Motion
Model
Cinematic quality · 720p · 5s · fal.ai
Output
🎬
Your generated video will appear here
Generating video...
History
Description
Mood Preset
Duration
Tempo
Volume Arc
Reference Track (optional)
🎶
Upload reference track
Output
Generated tracks will appear here.
History
Describe the Sound Effect
Category
Duration
Output
Generated sound effects will appear here.
History
🎤 Video Transcriber
Paste a YouTube or Vimeo URL and get a full text transcript. Works on any public video.
Transcript
🎚 Demucs: Audio Stem Separator
Isolates the voice from a mixed audio/video file. Use this BEFORE Wav2Lip if your video already has music mixed in — Wav2Lip needs a clean voice-only track.
🎵
📎 Upload video or audio (MP4, MOV, MP3, WAV)
Outputs all main stems, plus a quick vocals-only download when available.
👄 Wav2Lip: Lip Sync Fix
Requires a clean voice-only audio track (no music or SFX). If your audio is mixed, run Demucs first to isolate the vocals.
🎬
🎬 Upload AI-generated video (MP4, MOV)
🎙
🎙 Upload clean voice-only audio
🎛 Sound Combiner
Layer multiple audio tracks with individual volume control
Vol
Vol
Vol
Vol