Update app.py

app.py

@@ -59,10 +59,6 @@ def tts(text):
 
 
 def predict(image, prompt):
-    # text = captioner(image)[0]["generated_text"]
-
-    # audio_output = "output.wav"
-    # tts.tts_to_file(text, speaker=tts.speakers[0], language="en", file_path=audio_output)
 
     pixel_values = vqa_processor(images=image, return_tensors="pt").pixel_values
 
@@ -84,7 +80,10 @@ demo = gr.Interface(
     inputs=[gr.Image(type="pil",label="Environment"), gr.Textbox(label="Prompt", value="What is in the scene?")],
     outputs=[gr.Textbox(label="Caption"), gr.Audio(type="numpy",label="Audio Feedback")],
     css=".gradio-container {background-color: #002A5B}",
-    theme=gr.themes.Soft()
+    theme=gr.themes.Soft().set(
+        button_primary_background_fill="#AAAAAA",
+        button_primary_border="*button_primary_background_fill_dark"
+    )
 )
 
 demo.launch()