Update app.py

app.py

@@ -4,6 +4,7 @@ lpmc_client = gr.load("seungheondoh/LP-Music-Caps-demo", src="spaces")
 from gradio_client import Client
 
 client = Client("https://ysharma-explore-llamav2-with-tgi.hf.space/")
+sd = gr.load("runwayml/stable-diffusion-v1-5", src="spaces")
 
 from pydub import AudioSegment
 
@@ -47,7 +48,7 @@ def infer(audio_file):
     llama_q = f"""
 
     I'll give you music description, then i want you to provide an illustrative image description that would fit well with the music.
-    Answer with only one image description. Never do lists. Also specify in the description that we do not want photography.
+    Answer with only one image description. Never do lists.
 
     Here's the music description :
 
@@ -61,8 +62,10 @@ def infer(audio_file):
     )
     
     print(result)
+
+    image = sd(result, fn_index=0)[0]
     
-    return cap_result, result
+    return cap_result, result, image
 
 with gr.Blocks() as demo:
     with gr.Column(elem_id="col-container"):
@@ -72,6 +75,6 @@ with gr.Blocks() as demo:
         llama_trans_cap = gr.Textbox(label="Llama translation")
         img_result = gr.Video(label="Result")
 
-    infer_btn.click(fn=infer, inputs=[audio_input], outputs=[lpmc_cap, llama_trans_cap])
+    infer_btn.click(fn=infer, inputs=[audio_input], outputs=[lpmc_cap, llama_trans_cap, img_result])
 
 demo.queue().launch()