Spaces:
Sleeping
Sleeping
Update app.py
Browse files
app.py
CHANGED
@@ -164,8 +164,12 @@ def spectrogram_to_audio(magnitude_spectrogram):
|
|
164 |
return audio
|
165 |
|
166 |
|
167 |
-
# Function to generate audio from an uploaded image
|
168 |
def generate_audio_from_image(image):
|
|
|
|
|
|
|
|
|
|
|
169 |
test_img = image_transform(image).unsqueeze(0).to(device) # Preprocess image
|
170 |
|
171 |
# Generate sound spectrogram from the image using the loaded generator
|
@@ -178,6 +182,7 @@ def generate_audio_from_image(image):
|
|
178 |
# Convert audio tensor to numpy and return it for Gradio to handle
|
179 |
return generated_audio.numpy(), sample_rate
|
180 |
|
|
|
181 |
# Gradio Interface
|
182 |
def main():
|
183 |
global generator # Declare the generator object globally
|
@@ -190,8 +195,9 @@ def main():
|
|
190 |
|
191 |
# Gradio interface: allow users to upload an image and generate audio
|
192 |
iface = gr.Interface(fn=generate_audio_from_image,
|
193 |
-
|
194 |
-
|
|
|
195 |
|
196 |
iface.launch()
|
197 |
|
|
|
164 |
return audio
|
165 |
|
166 |
|
|
|
167 |
def generate_audio_from_image(image):
|
168 |
+
if image is None:
|
169 |
+
raise ValueError("The uploaded image is 'None'. Please check the Gradio input.")
|
170 |
+
|
171 |
+
# Ensure the image is in the right format
|
172 |
+
print(f"Image received: {type(image)}") # Debugging: Check if image is received
|
173 |
test_img = image_transform(image).unsqueeze(0).to(device) # Preprocess image
|
174 |
|
175 |
# Generate sound spectrogram from the image using the loaded generator
|
|
|
182 |
# Convert audio tensor to numpy and return it for Gradio to handle
|
183 |
return generated_audio.numpy(), sample_rate
|
184 |
|
185 |
+
|
186 |
# Gradio Interface
|
187 |
def main():
|
188 |
global generator # Declare the generator object globally
|
|
|
195 |
|
196 |
# Gradio interface: allow users to upload an image and generate audio
|
197 |
iface = gr.Interface(fn=generate_audio_from_image,
|
198 |
+
inputs=gr.Image(type="pil"), # PIL type image
|
199 |
+
outputs=gr.Audio(type="numpy", label="Generated Audio"))
|
200 |
+
|
201 |
|
202 |
iface.launch()
|
203 |
|