Clone any voice with AI — Create personalized voice synthesis from just 10-30 seconds of audio. Powered by Coqui XTTS.
Step 1: Upload 10s of audio → "This is my voice"
Step 2: AI clones the voice
Step 3: Generate anything → "Hello, I am cloned!"
Voice AI is exploding — from content creation to accessibility, voice cloning is revolutionizing communication.
| Feature | Description |
|---|---|
| 🎯 Voice Cloning | Clone from 10-30 seconds of audio |
| 🌍 Multi-language | English, Arabic, Spanish, French, German |
| ⚡ Fast | Generate speech in seconds |
| 🔊 High Quality | 44.1kHz output |
| 🎛️ Gradio UI | Beautiful web interface |
| 🐳 Docker | One-command deployment |
git clone https://github.com/walidsobhie-code/ai-voice-clone.git
cd ai-voice-clone
pip install -r requirements.txt# Upload 10-30 seconds of clear audio
python clone_voice.py --input my_voice.wav --output my_cloned_voice
# Output:
# 🎤 Loading audio: my_voice.wav
# 🔄 Cloning voice...
# ✅ Voice cloned successfully!python synthesize.py --model my_cloned_voice.wav \
--text "Hello everyone, this is my cloned voice!" \
--output hello.wavpython gradio_app.py
# Opens: http://localhost:7860┌─────────────────────────────────────────────────────────┐
│ 🎙️ AI Voice Clone │
├─────────────────────────────────────────────────────────┤
│ │
│ [🎤 Clone Voice] [🎤 Generate Speech] │
│ │
│ ┌──────────────────┐ ┌──────────────────┐ │
│ │ Upload Audio: │ │ Text Input: │ │
│ │ [my_voice.wav] │ │ Hello world! │ │
│ │ Duration: 15s │ │ Speed: [1.0x] │ │
│ └──────────────────┘ └──────────────────┘ │
│ │
│ [🔄 Clone Voice] [🎤 Generate] │
│ │
│ Status: │
│ ✅ Voice cloned successfully! │
│ 📊 Model: my_cloned_voice │
└─────────────────────────────────────────────────────────┘
from clone_voice import clone_voice
from synthesize import synthesize
# Step 1: Clone voice
result = clone_voice(
input_file="my_voice.wav",
output_name="my_model"
)
print(result)
# {'status': 'success', 'output': 'my_model.wav'}
# Step 2: Generate speech
result = synthesize(
model_path="my_model.wav",
text="Hello, I sound exactly like the original!",
output="output.wav"
)
print(result)
# {'status': 'success', 'output': 'output.wav'}| Industry | Use Case |
|---|---|
| 🎬 Content | Create videos with any voice |
| 🎧 Podcast | Clone voices for narration |
| ♿ Accessibility | Read text in any voice |
| 🏢 Brand | Consistent brand voice across content |
| 📚 Education | Localize content in local voices |
| 🎮 Gaming | Character voices for games |
Original Audio (10-30s)
↓
XTTS Encoder (extracts voiceprint)
↓
Latent Space Representation
↓
XTTS Decoder (generates new audio)
↓
Your Cloned Voice Saying Anything!
# Build
docker build -t voice-clone .
# Run
docker run -p 7860:7860 voice-cloneai-voice-clone/
├── clone_voice.py # Voice cloning
├── synthesize.py # Speech synthesis
├── gradio_app.py # Web UI
├── requirements.txt
├── Dockerfile
└── examples/
├── basic_clone.py
└── multi_language.py
This tool should only be used ethically:
- ✅ With consent of the voice owner
- ✅ For legitimate purposes (accessibility, entertainment)
- ❌ NOT for fraud, impersonation, or deception
See CONTRIBUTING.md
If this helped you, please star the repo!
Built with ❤️ by walidsobhie-code
- [Planned] Web version / hosted demo
- [Planned] API endpoint for production use
- [Planned] Support for more languages
- [In Progress] Performance optimizations
- [Done] Gradio web interface
- [Done] Docker deployment
Have a project using this? Send a PR to add your company!
- (coming soon — be the first to list your project!)
We welcome contributions! Please see CONTRIBUTING.md for guidelines.