Update app.py

app.py

@@ -1,49 +1,79 @@
+# app.py
+# =============
+# This is a complete app.py file for a Gradio application that allows users to upload an audio file and generate a video with frequency visualization.
+
 import gradio as gr
-import librosa
-import librosa.display
 import numpy as np
 import matplotlib.pyplot as plt
-from moviepy.editor import VideoClip, AudioFileClip, CompositeVideoClip
+import librosa
+import librosa.display
+import cv2
+import os
+import moviepy.video.io.ImageSequenceClip
 
-def generate_frequency_visualization(audio_file):
+# Function to generate frequency visualization frames from audio
+def generate_frequency_visualization(audio_path):
     # Load the audio file
-    y, sr = librosa.load(audio_file, sr=None)
+    y, sr = librosa.load(audio_path)
 
-    # Compute the Short-Time Fourier Transform (STFT)
+    # Perform Short-Time Fourier Transform (STFT)
     D = librosa.amplitude_to_db(np.abs(librosa.stft(y)), ref=np.max)
 
-    # Create a figure for the visualization
-    fig, ax = plt.subplots()
-    img = librosa.display.specshow(D, sr=sr, ax=ax, y_axis='linear', fmax=8000)
-    fig.colorbar(img, ax=ax, format="%+2.0f dB")
-    ax.set(title='Frequency Visualization')
-    plt.axis('off')
-    plt.savefig('frequency_visualization.png', bbox_inches='tight', pad_inches=0, dpi=100)
-    plt.close(fig)
+    # Create a directory to save the frames
+    os.makedirs('frames', exist_ok=True)
 
-    # Load the audio file
-    audio_clip = AudioFileClip(audio_file)
+    # Generate and save each frame
+    for i, frame in enumerate(D.T):
+        plt.figure(figsize=(10, 6))
+        librosa.display.specshow(frame.reshape(1, -1), sr=sr, x_axis='time', y_axis='log')
+        plt.axis('off')
+        plt.savefig(f'frames/frame_{i:04d}.png', bbox_inches='tight', pad_inches=0)
+        plt.close()
 
-    # Create a video clip from the frequency visualization image
-    video_clip = VideoClip(lambda t: plt.imread('frequency_visualization.png'), duration=audio_clip.duration)
+    return 'frames'
 
-    # Combine the audio and video clips
-    final_clip = video_clip.set_audio(audio_clip)
+# Function to create a video from the generated frames
+def create_video_from_frames(frames_directory):
+    # Get the list of frame files
+    frame_files = [os.path.join(frames_directory, f) for f in os.listdir(frames_directory) if f.endswith('.png')]
+    frame_files.sort()
 
-    # Write the final video to a file
-    output_file = 'frequency_visualization.mp4'
-    final_clip.write_videofile(output_file, codec='libx264', audio_codec='aac')
+    # Create a video from the frames
+    clip = moviepy.video.io.ImageSequenceClip.ImageSequenceClip(frame_files, fps=30)
+    video_path = 'output_video.mp4'
+    clip.write_videofile(video_path, codec='libx264')
 
-    return output_file
+    return video_path
+
+# Gradio interface function
+def process_audio(audio):
+    audio_path = audio
+    frames_directory = generate_frequency_visualization(audio_path)
+    video_path = create_video_from_frames(frames_directory)
+    return video_path
 
 # Create the Gradio interface
 iface = gr.Interface(
-    fn=generate_frequency_visualization,
-    inputs=gr.Audio(source="upload", type="file"),
-    outputs=gr.Video(label="Frequency Visualization Video"),
+    fn=process_audio,
+    inputs=gr.Audio(source="upload", type="filepath"),
+    outputs=gr.Video(label="Generated Video"),
     title="Audio Frequency Visualization",
     description="Upload an audio file to generate a video with frequency visualization."
 )
 
 # Launch the Gradio interface
-iface.launch()
+if __name__ == "__main__":
+    iface.launch()
+
+# Dependencies
+# =============
+# The following dependencies are required to run this app:
+# - librosa
+# - numpy
+# - matplotlib
+# - opencv-python
+# - moviepy
+# - gradio
+#
+# You can install these dependencies using pip:
+# pip install librosa numpy matplotlib opencv-python moviepy gradio