create_test_audio.py

#!/usr/bin/env python3
"""
Create a test audio file for testing the MPC Speech service
"""

import numpy as np
import soundfile as sf
from pathlib import Path

def create_test_audio(filename="test_audio.wav", duration=3.0, sample_rate=16000):
    """
    Create a simple test audio file with a sine wave tone
    
    Args:
        filename: Output filename
        duration: Duration in seconds
        sample_rate: Sample rate in Hz
    """
    print(f"🎵 Creating test audio file: {filename}")
    
    # Generate time array
    t = np.linspace(0, duration, int(sample_rate * duration), False)
    
    # Create a simple tone (440 Hz - A note) with some variation
    frequency1 = 440  # A note
    frequency2 = 523  # C note
    
    # Create a chord-like sound
    audio = (
        0.3 * np.sin(2 * np.pi * frequency1 * t) +  # A note
        0.2 * np.sin(2 * np.pi * frequency2 * t) +  # C note
        0.1 * np.sin(2 * np.pi * 660 * t)          # E note
    )
    
    # Add some envelope to make it sound more natural
    envelope = np.exp(-t * 0.5)  # Exponential decay
    audio = audio * envelope
    
    # Normalize to prevent clipping
    audio = audio / np.max(np.abs(audio)) * 0.8
    
    # Save as WAV file
    sf.write(filename, audio, sample_rate)
    
    file_size = Path(filename).stat().st_size
    print(f"✅ Created {filename}")
    print(f"   Duration: {duration}s")
    print(f"   Sample rate: {sample_rate} Hz")
    print(f"   File size: {file_size:,} bytes")
    
    return filename

def create_silent_audio(filename="silent_audio.wav", duration=2.0, sample_rate=16000):
    """Create a silent audio file for testing"""
    print(f"🔇 Creating silent audio file: {filename}")
    
    # Create silent audio (zeros)
    audio = np.zeros(int(sample_rate * duration))
    
    # Save as WAV file
    sf.write(filename, audio, sample_rate)
    
    file_size = Path(filename).stat().st_size
    print(f"✅ Created {filename}")
    print(f"   Duration: {duration}s (silent)")
    print(f"   Sample rate: {sample_rate} Hz")
    print(f"   File size: {file_size:,} bytes")
    
    return filename

def create_speech_like_audio(filename="speech_test.wav", duration=5.0, sample_rate=16000):
    """Create a more speech-like test audio with varying frequencies"""
    print(f"🎤 Creating speech-like audio file: {filename}")
    
    # Generate time array
    t = np.linspace(0, duration, int(sample_rate * duration), False)
    
    # Create speech-like formants (simplified)
    # Human speech typically has formants around 500Hz, 1500Hz, 2500Hz
    formant1 = 500
    formant2 = 1500
    formant3 = 2500
    
    # Create a more complex waveform
    audio = (
        0.4 * np.sin(2 * np.pi * formant1 * t) * (1 + 0.3 * np.sin(2 * np.pi * 3 * t)) +
        0.3 * np.sin(2 * np.pi * formant2 * t) * (1 + 0.2 * np.sin(2 * np.pi * 5 * t)) +
        0.2 * np.sin(2 * np.pi * formant3 * t) * (1 + 0.1 * np.sin(2 * np.pi * 7 * t))
    )
    
    # Add some noise to make it more realistic
    noise = 0.05 * np.random.normal(0, 1, len(audio))
    audio = audio + noise
    
    # Create segments (like words)
    segment_duration = 0.8
    pause_duration = 0.2
    segment_samples = int(segment_duration * sample_rate)
    pause_samples = int(pause_duration * sample_rate)
    
    # Apply segmentation
    for i in range(0, len(audio), segment_samples + pause_samples):
        # Keep segment
        segment_end = min(i + segment_samples, len(audio))
        # Add pause
        pause_start = segment_end
        pause_end = min(pause_start + pause_samples, len(audio))
        if pause_end > pause_start:
            audio[pause_start:pause_end] *= 0.1  # Reduce volume for pause
    
    # Normalize
    audio = audio / np.max(np.abs(audio)) * 0.7
    
    # Save as WAV file
    sf.write(filename, audio, sample_rate)
    
    file_size = Path(filename).stat().st_size
    print(f"✅ Created {filename}")
    print(f"   Duration: {duration}s")
    print(f"   Sample rate: {sample_rate} Hz")
    print(f"   File size: {file_size:,} bytes")
    print(f"   Note: This is synthetic audio for testing")
    
    return filename

def main():
    """Create test audio files"""
    print("🎧 Creating test audio files for MPC Speech service")
    print("=" * 50)
    
    try:
        # Create different types of test audio
        create_test_audio("audio.wav", duration=3.0)
        create_silent_audio("silent_test.wav", duration=2.0)
        create_speech_like_audio("speech_test.wav", duration=4.0)
        
        print(f"\n✅ All test audio files created successfully!")
        print(f"📁 Files created in current directory:")
        print(f"   • audio.wav - Simple tone (for basic testing)")
        print(f"   • silent_test.wav - Silent audio (edge case testing)")
        print(f"   • speech_test.wav - Speech-like audio (more realistic)")
        print(f"\n🧪 You can now test with: python testClient.py")
        
    except Exception as e:
        print(f"❌ Error creating test audio files: {e}")
        print(f"💡 Make sure you have soundfile installed: pip install soundfile")

if __name__ == "__main__":
    main()