Voice Cloning

Voice Cloning Guide

Clone any voice using our advanced VAL v1 model to create custom AI voice characters. This guide covers all key aspects of the voice cloning process.

Cloning Methods

Method 1: File Upload

Upload pre-recorded audio files to create voice characters.

Supported Formats:

File Requirements:

Duration: 10-30 seconds optimal
Quality: Clear audio with minimal background noise
Size: Maximum 10MB per file
Sample Rate: 16kHz or higher recommended

Steps:

Visit the Voice Clone page
Click the "Upload File" tab
Select your audio file
Wait for upload to complete
Proceed to character configuration

Method 2: Online Recording

Record audio directly in your browser for instant cloning.

Requirements:

Microphone access permission
Quiet recording environment
Stable internet connection

Steps:

Visit the Voice Clone page
Click the "Online Recording" tab
Allow browser microphone access
Click record button to start
Speak clearly for 10-30 seconds
Click stop when finished
Preview your recording
Proceed to character configuration

Character Configuration

After providing audio input, configure your character details:

Required Fields

Character Name

Give your voice character a memorable name
Use descriptive names for easy identification
Maximum 20 characters

Optional Fields

Character Description

Describe voice characteristics
Note intended use case
Add relevant context
Maximum 200 characters

Character Avatar

Upload profile image for visual identification
Supported formats: JPG, PNG, WebP
Maximum size: 5MB
Recommended dimensions: 512x512 pixels

Advanced Settings

Model Version

Select the AI model for cloning:

VAL v1 (Recommended): Latest model with best quality and multilingual support

Language Selection

Choose the primary language of your audio sample:

English (US, UK, Australian)
Chinese (Mandarin, Cantonese)
Spanish, French, German, Italian
Japanese, Korean
And 20+ more languages

Tip: Selecting the correct language improves cloning accuracy.

Vocal Separation

Enable this option if your audio contains background noise or music:

Enabled: AI will isolate and extract vocals from background sounds
Disabled: Use for clean audio recordings

When to Enable:

Audio has background music
Multiple speakers in recording
Environmental noise present
Phone call recordings

When to Disable:

Studio-quality recordings
Clean voice-only audio
Minimal background noise

Cloning Process

After clicking "Create Voice Character":

Upload: Audio is securely uploaded to our servers
Processing: VAL v1 model analyzes voice characteristics
Training: AI learns unique vocal patterns
Validation: Quality checks ensure successful cloning
Completion: Character is ready for use

Processing Time:

Short samples (10-30s): 5-15 seconds
Medium samples (1-5 min): 15-30 seconds
Long samples (5+ min): 30-60 seconds

Quality Tips

For best cloning results:

Audio Quality

✅ Use high-quality microphones or recordings
✅ Record in quiet environments
✅ Avoid echo and reverb
✅ Maintain consistent volume levels
❌ Avoid compressed or low-bitrate audio

Content Selection

✅ Include varied sentences and expressions
✅ Capture different emotional tones
✅ Use natural speaking pace
✅ Include pauses and breathing
❌ Avoid monotone or robotic speech

Sample Length

Recommended (10-30s): Quick cloning with quality assurance

Troubleshooting

Common Issues

"Audio quality too low"

Solution: Use higher quality recording equipment or format
Ensure sample rate is at least 16kHz

"Background noise detected"

Solution: Enable vocal separation option
Re-record in quieter environment

"Sample too short"

Solution: Provide at least 10 seconds of clear speech
Combine multiple short clips if needed

"Language mismatch"

Solution: Ensure selected language matches audio content
Try "Auto-detect" if unsure

Usage Limits

Character creation is subject to your plan limits:

Plan	Characters	Monthly Clones
Starter	10	10
Standard	100	100
Premium	Unlimited	Unlimited

Subject to change, please refer to the Pricing page for current rates.

Check your current usage in the user menu or My Voices page.

Next Steps

After creating your character:

Preview: Test the voice in Text-to-Speech
Organize: Manage characters in Character Management
Share: Publish to Voice Square (optional)
Create: Use in your projects and content

Need help? Check our FAQ or contact [email protected]