Powered by MiniMax TTS + SadTalker via FAL.ai
Make any photo speak
your script
Upload a portrait photo, type your script, and our AI generates a realistic talking video with lip-sync and voice — in any language including Vietnamese.
📸 Portrait Photo
👤
Upload a portrait
Face should be clearly visible
JPG, PNG or WebP
JPG, PNG or WebP
📝 Script to Read
Type or paste the script
0 / 1000
✦ Omnihuman 1.5 + MiniMax TTS · Real MP4 with audio
⚙️ Generating your talking video…
1
Uploading portrait photo
Uploading to FAL CDN…
2
Converting script to speech (MiniMax TTS)
Waiting…
3
Audio ready — no base video needed
Waiting…
4
Generating talking video (SadTalker)
Waiting… (~1–2 min)
Preparing…
Features
How it works
🗣️
Script to Speech
MiniMax TTS converts your script into natural-sounding speech in Vietnamese, English, Spanish and more.
👄
Realistic Lip Sync
Omnihuman 1.5 animates the portrait with precise lip movements, facial expressions and head motion matching the audio.
🌐
Any Language
Supports Vietnamese, English, Spanish, Chinese and many more languages for the script reading.
🎬
Real MP4 with Audio
Downloads as a proper MP4 video file with synchronized audio — ready to share on any platform.