Update app.py

app.py

@@ -178,15 +178,15 @@ def generate_audio_from_image(image):
     # Transpose to (samples, channels) for stereo
     generated_audio = generated_audio.T
 
-    # Ensure that the audio is a 2D array (for stereo)
+    # Ensure the audio is 2D (even if mono)
     if generated_audio.ndim == 1:
         generated_audio = np.expand_dims(generated_audio, axis=-1)  # Add channel dimension for mono
 
     # Debug: Print the shape and type of the generated audio
     print(f"Generated audio shape after processing: {generated_audio.shape}, type: {generated_audio.dtype}")
 
-    # Return the audio and the sample rate (ensure sample rate is an integer)
-    return generated_audio, int(sample_rate)
+    # Explicit return of the tuple (audio_data, sample_rate)
+    return generated_audio, sample_rate