XTTS-v2 - Voice Cloning

By accessing or using any feature within this space, you acknowledge and accept the terms of the following license: https://coqui.ai/cpml.

Model source: coqui/XTTS-v2

⚠️ Disclaimer and Legal Notice

By using this voice cloning application, you acknowledge and agree to the following:

  1. This application is provided "as is" without any warranties of any kind, either express or implied.
  2. The creator(s) of this application accept no responsibility or liability for any misuse of the technology.
  3. You are solely responsible for obtaining proper consent when cloning someone else's voice.
  4. You agree not to use this technology for deceptive, harmful, or illegal purposes.
  5. Voice cloning results may vary in quality and accuracy; no specific results are guaranteed.
  6. You understand that voice cloning technology has ethical implications and agree to use it responsibly.

The technology is intended for legitimate creative, educational, and accessibility purposes only.

🎯 Quick User Guide

📝 In Summary

  1. Choose a reference audio (example or upload)
  2. Enter your text and select the language
  3. Click on "Generate Audio"
  4. If needed, regenerate multiple times or adjust parameters

🔍 Essential Tips

Reference Audio Quality

  • Generated audio quality directly depends on your reference
  • Use a clear recording, without background noise (3-10 seconds)
  • If the result is not satisfactory, try another reference

Text and Pronunciation

  • Automatic preprocessing improves pronunciation
  • For long texts, native XTTS splitting is recommended
  • Text analysis detects problematic elements (URLs, symbols, etc.)

Optimizing Results

  • Regenerate multiple times: Each generation is unique (temperature)
  • Adjust speed for a more natural flow
  • For excessive silences, check silence removal options
  • Parameters are preset to values recommended by Coqui

Supported Languages

  • 17 languages available, including English, French, Spanish, etc.
  • Each language has its own pronunciation rules

Customization

  • Create custom replacement rules for specific cases
Language
0.1 1.5
0.5 2
Text Splitting Method

For long texts, use the recommended native XTTS splitting

1 5
1 2
10 50
0 50
0 1
-60 -20
300 1000
100 500
Apply to language
Reference Audio (examples)

Try different reference voices if the result doesn't suit you.