Update app.py

app.py

@@ -7,43 +7,71 @@ import numpy as np
 import matplotlib.pyplot as plt
 import librosa
 import librosa.display
-import cv2
 import os
 import moviepy.video.io.ImageSequenceClip
 
 # Function to generate frequency visualization frames from audio
 def generate_frequency_visualization(audio_path):
-    # Load the audio file
-    y, sr = librosa.load(audio_path)
+    try:
+        # Load the audio file
+        y, sr = librosa.load(audio_path, sr=None)
 
-    # Perform Short-Time Fourier Transform (STFT)
-    D = librosa.amplitude_to_db(np.abs(librosa.stft(y)), ref=np.max)
+        if sr == 0 or len(y) == 0:
+            raise ValueError("Invalid audio file: sampling rate or audio data is zero.")
 
+        # Perform Short-Time Fourier Transform (STFT)
+        D = librosa.amplitude_to_db(np.abs(librosa.stft(y)), ref=np.max)
+
+        # Create a directory to save the frames
+        os.makedirs('frames', exist_ok=True)
+
+        # 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()
+
+        return 'frames'
+    except Exception as e:
+        print(f"Error generating frequency visualization: {e}")
+        # Fallback: Generate a default visualization
+        generate_default_visualization()
+        return 'frames'
+
+# Function to generate a default visualization
+def generate_default_visualization():
     # Create a directory to save the frames
     os.makedirs('frames', exist_ok=True)
 
-    # Generate and save each frame
-    for i, frame in enumerate(D.T):
+    # Generate and save default frames
+    for i in range(10):  # Generate 10 default frames
         plt.figure(figsize=(10, 6))
-        librosa.display.specshow(frame.reshape(1, -1), sr=sr, x_axis='time', y_axis='log')
+        plt.plot(np.sin(np.linspace(0, 10, 100)) * (i + 1))
         plt.axis('off')
         plt.savefig(f'frames/frame_{i:04d}.png', bbox_inches='tight', pad_inches=0)
         plt.close()
 
-    return 'frames'
-
 # 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()
+    try:
+        # 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()
 
-    # 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')
+        if not frame_files:
+            raise ValueError("No frames found to create the video.")
 
-    return video_path
+        # 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 video_path
+    except Exception as e:
+        print(f"Error creating video from frames: {e}")
+        return None
 
 # Gradio interface function
 def process_audio(audio):
@@ -52,13 +80,20 @@ def process_audio(audio):
     video_path = create_video_from_frames(frames_directory)
     return video_path
 
-# Create the Gradio interface
+# Create the Gradio interface with explanations and recommendations
 iface = gr.Interface(
     fn=process_audio,
-    inputs=gr.Audio(type="filepath"),
+    inputs=gr.Audio(type="filepath", label="Upload Audio File"),
     outputs=gr.Video(label="Generated Video"),
     title="Audio Frequency Visualization",
-    description="Upload an audio file to generate a video with frequency visualization."
+    description="Upload an audio file to generate a video with frequency visualization. "
+                "Supported file types: WAV, MP3, FLAC. "
+                "Recommended file duration: 10 seconds to 5 minutes. "
+                "If the file is invalid or cannot be processed, a default visualization will be generated.",
+    examples=[
+        ["examples/sample_audio.wav"],
+        ["examples/sample_audio.mp3"]
+    ]
 )
 
 # Launch the Gradio interface
@@ -71,9 +106,8 @@ if __name__ == "__main__":
 # - librosa
 # - numpy
 # - matplotlib
-# - opencv-python
 # - moviepy
 # - gradio
 #
 # You can install these dependencies using pip:
-# pip install librosa numpy matplotlib opencv-python moviepy gradio
+# pip install librosa numpy matplotlib moviepy gradio