✅ Happy with the voice? Click Generate Video below.
🎵
Click to upload audio file
WAV, MP3, M4A supported
Advanced Options
Loading backgrounds…
Auto Captions / Subtitles
Caption Style
Apply Brand Kit
🎬 Stabilize Video (fixes shaky lips)
Translation
🎨 Quality
Processing…
Preview
🎥
Your video will appear here
🎬 —⏱ Generated in —
Recent Videos
No videos yet
My Cloned Voices
No cloned voices yet. Add your own!
Stock Voices (Kokoro TTS)
Loading…
Backgrounds
Loading…
Background Music
Loading…
Brand Kit
🏢
Upload your logo (PNG with transparency recommended)
Intro / Outro Clips
⏮
Upload intro (MP4)
⏭
Upload outro (MP4)
Account
Loading…
Upload Avatar
📷
Photo
JPG/PNG
🎬
Video
MP4/MOV 5–30s
📷 Upload a clear, front-facing face photo. Head should fill most of the frame.
📷
Click to browse or drag & drop
Clone Your Voice
Record or upload 30–60 seconds of your voice speaking clearly.
XTTS v2 will clone it instantly — no API keys, runs entirely on your server.
Click to start recording
0:00
🎵
Upload audio files (WAV/MP3/M4A)
Select multiple files for better accuracy
Voice Tuning
Speed1.0×
SlowerFaster
Accent Consistency0.65
More consistent (less accent)More expressive
💡 For best results: upload 3–5 clips of 20–30s each in a quiet room, speaking clearly and naturally.
Lower accent consistency reduces accent drift from the model's training data.
Tune Voice —
Adjust the sliders, click Preview to hear the result,
then Save when happy.
Your original recordings are preserved — you can re-tune anytime.
Speed1.0×
SlowerFaster
Accent Consistency0.65
More consistent / less accentMore expressive
Preview output:
Tune Avatar —
These settings control how MuseTalk positions the lip-sync region for this specific avatar.
Adjust, generate a test video, and save when the mouth looks natural.
