Music AI
🎤 Voice Cloning Guide
Clone any voice from audio samples and use it in your AI-generated music.
15 min
Intermediate
For best voice cloning results, your audio samples should meet these criteria:
⏱️
30-60 seconds
Of clean vocal audio
🎙️
Solo Voice Only
No music or background
📊
High Quality
WAV/FLAC, 44.1kHz+
🔇
Low Noise
Minimal background noise
Ideal sample: Clear speech or singing, no reverb, consistent volume
Acceptable: Some room echo, minor background noise
Poor quality: Music in background, heavy reverb, multiple voices
- Open Music AI in Vigthoria Coder
- Navigate to Voice Library → Create Voice
- Click "Upload Audio" or drag and drop your file
- Give your voice a descriptive name (e.g., "Deep Male Baritone")
- Add optional tags for easier searching later
Upload multiple samples of the same voice for better cloning accuracy. The AI will learn from all of them to create a more accurate voice model.
After uploading, Music AI will:
- Analyze the audio for voice characteristics
- Extract vocal features (pitch, timbre, cadence)
- Train a voice model (takes 2-5 minutes)
- Generate a preview sample
You'll receive a notification when your voice is ready.
Your new voice appears in the Voice Library. To use it:
- Open the Voice Selection panel when generating a song
- Filter by "My Voices"
- Select your cloned voice
- Generate your song as usual
The cloned voice will sing your lyrics with the characteristics of the original audio sample.
If your voice clone doesn't sound right:
- Add more samples — Upload additional audio with varied content
- Clean the audio — Remove background noise before uploading
- Match the style — Use singing samples for singing, speech for speech
- Check pitch range — Ensure your lyrics match the voice's natural range
Only clone voices you have permission to use. Voice cloning of public figures, celebrities, or other people without consent violates our Terms of Service and may be illegal.