scripts/finetune_xtts.py

#!/usr/bin/env python3
"""
Fine-tune XTTS v2 for Nigerian languages (Yoruba, Hausa, Igbo, Pidgin).
Uses Coqui TTS framework.
"""
import os
import sys
from pathlib import Path

# Check for GPU
import torch
print(f"PyTorch version: {torch.__version__}")
print(f"CUDA available: {torch.cuda.is_available()}")
if torch.cuda.is_available():
    print(f"GPU: {torch.cuda.get_device_name(0)}")
    print(f"GPU Memory: {torch.cuda.get_device_properties(0).total_memory / 1024**3:.1f} GB")

BASE_DIR = Path.home() / "voice-training"
PREPARED_DIR = BASE_DIR / "prepared_data"
OUTPUT_DIR = BASE_DIR / "output"

def setup_xtts_training():
    """Setup XTTS v2 fine-tuning."""
    try:
        from TTS.tts.configs.xtts_config import XttsConfig
        from TTS.tts.models.xtts import Xtts
        print("XTTS modules imported successfully")
        return True
    except ImportError as e:
        print(f"Error importing TTS: {e}")
        print("Installing coqui-tts...")
        os.system("pip install coqui-tts")
        return False

def main():
    print("=== XTTS v2 Fine-tuning for Nigerian Languages ===")
    print(f"Base directory: {BASE_DIR}")
    
    if not setup_xtts_training():
        print("Please install dependencies and try again")
        sys.exit(1)
    
    # Check for prepared data
    if not PREPARED_DIR.exists():
        print(f"Prepared data not found at {PREPARED_DIR}")
        print("Run prepare_data.py first")
        sys.exit(1)
    
    print("Ready for fine-tuning!")
    print("\nNext steps:")
    print("1. Ensure audio data is prepared")
    print("2. Configure training parameters")
    print("3. Start training")

if __name__ == "__main__":
    main()