Update app.py

app.py

@@ -40,14 +40,14 @@ def speech_to_text(mic=None, file=None):
 
 def zero_shot(image, text_input):
     results = zero(image, text_input)
-    return {dic["label"]: dic["score"] for dic in results}
+    return results
     
 with gr.Blocks() as demo: 
     gr.Markdown( """
-    ## Input voice/text 
-    ## Convert voice/text to image via dalle
-    ## Given list of labels and a selected image from gallery do zero-shot classification 
-    ## Coming soon: TTS(audio) your output label as: Your output looks like "label of zero-shot"
+    Input voice/text 
+    Convert voice/text to image via Latent Diffusion
+    Given list of labels and a selected image from gallery do zero-shot classification 
+    Coming soon: TTS(audio) your output label as: Your output looks like "label of zero-shot"
      """)    
     with gr.Row():
         with gr.Column():