Update app.py

app.py

@@ -1,18 +1,20 @@
 import gradio as gr
-from transformers import CLIPProcessor, CLIPModel, pipeline
+from transformers import CLIPProcessor, CLIPModel, pipeline, AutoProcessor, MusicgenForConditionalGeneration
 import torch
 from PIL import Image
 import scipy.io.wavfile
 
 # Load the MusicGen model
-
+#musicgen = pipeline("text-to-audio", model="facebook/musicgen-small")
+musicProcessor = AutoProcessor.from_pretrained("facebook/musicgen-small")
+musicgen = MusicgenForConditionalGeneration.from_pretrained("facebook/musicgen-small")
 # Load the StreetCLIP model
 model = CLIPModel.from_pretrained("geolocal/StreetCLIP")
 processor = CLIPProcessor.from_pretrained("geolocal/StreetCLIP")
 
 labels = ['Albania', 'Andorra', 'Argentina', 'Australia', 'Austria', 'Bangladesh', 'Belgium', 'Bermuda', 'Bhutan', 'Bolivia', 'Botswana', 'Brazil', 'Bulgaria', 'Cambodia', 'Canada', 'Chile', 'China', 'Colombia', 'Croatia', 'Czech Republic', 'Denmark', 'Dominican Republic', 'Egypt', 'Ecuador', 'Estonia', 'Finland', 'France', 'Germany', 'Ghana', 'Greece', 'Greenland', 'Guam', 'Guatemala', 'Hungary', 'Iceland', 'India', 'Indonesia', 'Ireland', 'Israel', 'Italy', 'Japan', 'Jordan', 'Kenya', 'Kyrgyzstan', 'Laos', 'Latvia', 'Lesotho', 'Lithuania', 'Luxembourg', 'Macedonia', 'Madagascar', 'Malaysia', 'Malta', 'Mexico', 'Monaco', 'Mongolia', 'Montenegro', 'Netherlands', 'New Zealand', 'Nigeria', 'Norway', 'Pakistan', 'Palestine', 'Peru', 'Philippines', 'Poland', 'Portugal', 'Puerto Rico', 'Romania', 'Russia', 'Rwanda','Saudi Arabia', 'Senegal', 'Serbia', 'Singapore', 'Slovakia', 'Slovenia', 'South Africa', 'South Korea', 'Spain', 'Sri Lanka', 'Swaziland', 'Sweden', 'Switzerland', 'Syria','Taiwan', 'Thailand', 'Tunisia', 'Turkey', 'Uganda', 'Ukraine', 'United Arab Emirates', 'United Kingdom', 'United States', 'Uruguay']
 
-def process_image(image):
+def process_image(image, audio_path="musicgen_out.wav"):
     # Ensure the image is in the correct format
     if isinstance(image, str):
         image = Image.open(image)
@@ -28,19 +30,29 @@ def process_image(image):
 
     # Get the country with the highest probability
     country_index = probs.argmax(dim=1).item()
-    print(country_index)
     country = labels[country_index]
-    print(country)
+
     # Generate music based on the country
+    music_description = f"Traditional music from {country}"
+    #music = musicgen(music_description, forward_params={"do_sample": True})
+    inputs = musicProcessor(
+    text=[music_description],
+    padding=True,
+    return_tensors="pt",
+    )
+    audio_values = musicgen.generate(**inputs, max_new_tokens=256)
+
 
     # Save the generated music to the specified path
+    sampling_rate = model.config.audio_encoder.sampling_rate
+    scipy.io.wavfile.write("musicgen_out.wav", rate=sampling_rate, data=audio_values[0, 0].numpy())
 
     # Return the country and the path to the generated music
-    return country
+    return country, audio_path
 
 # Define the Gradio interface
 inputs = gr.Image(type="pil", label="Upload a photo (تحميل صورة)")
-outputs = [gr.Textbox(label="Country (البلد)")]
+outputs = [gr.Textbox(label="Country (البلد)"), gr.Audio(label="Generated Music (الموسيقى المولدة)")]
 
 iface = gr.Interface(
     fn=process_image,
@@ -48,8 +60,8 @@ iface = gr.Interface(
     outputs=outputs,
     title="Photo to Country and Music Generator محدد الموقع من الصور بالاضافة الى انشاء م",
     description="Upload a photo to identify the country and generate traditional music from that country. (قم بتحميل صورة لتحديد البلد وإنشاء موسيقى تقليدية من هذا البلد.)",
-    examples=["Egypt.jfif", "Riyadh.jpeg", "Syria.jfif", "Turkey.jfif"]
+    examples=["/content/Egypt.jfif", "/content/Riyadh.jpeg", "/content/Syria.jfif", "/content/Turkey.jfif"]
 )
 
 # Launch the interface
-iface.launch()
+iface.launch(debug=True)