visualize-audio.py

import subprocess
import sys

import numpy as np
import matplotlib.pyplot as plt
from scipy.io import wavfile

def create_visualization(audio_file, output_image):
    # Load the audio file
    samplerate, data = wavfile.read(audio_file)
    if data.ndim > 1:  # Stereo
        data = data.mean(axis=1)  # Convert to mono by averaging channels

    # Perform a Fourier transform to get frequency spectrum
    freqs = np.fft.rfftfreq(len(data), 1/samplerate)
    fft_spectrum = np.abs(np.fft.rfft(data))

    # Simplified visualization: plot a logarithmic spectrum
    plt.figure(figsize=(10, 4))
    plt.semilogx(freqs, fft_spectrum)
    plt.xlabel('Frequency (Hz)')
    plt.ylabel('Amplitude')
    plt.title('Audio Spectrum')
    plt.savefig(output_image)
    plt.close()

def extract_audio(video_file, audio_file):
    subprocess.call(['ffmpeg', '-i', video_file, '-q:a', '0', '-map', 'a', audio_file, '-y'])

def combine_video_audio(original_video, visualization_video, output_file):
    subprocess.call(['ffmpeg', '-i', original_video, '-i', visualization_video, '-filter_complex', "[0:v][1:v] overlay=0:0", '-c:v', 'libx264', '-crf', '18', '-preset', 'veryfast', output_file, '-y'])

if __name__ == "__main__":
    if len(sys.argv) < 2:
        print("Usage: python script.py <input_video.mp4>")
        sys.exit(1)

    input_video = sys.argv[1]
    audio_file = "extracted_audio.mp3"
    visualization_video = "visualization.mp4"
    output_video = "output_with_visualization.mp4"

    # Step 1: Extract the audio from the video
    print("Extracting audio...")
    extract_audio(input_video, audio_file)

    # Step 2: Generate a visualization for the audio
    # Note: This step is not implemented here. You should replace the next print statement
    # with your method of generating a visualization video file named "visualization.mp4"
    print("Generating audio visualization... (This step is not implemented in the script)")

    # Assuming "visualization.mp4" is created and exists.

    # Step 3: Combine the original video with the audio visualization
    print("Combining video with visualization...")
    combine_video_audio(input_video, visualization_video, output_video)

    print("Process completed. The output video is:", output_video)