session state fixed

app.py

@@ -557,7 +557,7 @@ ADAPTIVE_PROMPT_TEMPLATES = [
     "A realistic video of a person speaking and sometimes looking directly to the camera and moving their eyes and pupils and head accordingly and turning and looking at the camera and looking away from the camera based on their movements with dynamic and rhythmic and extensive hand gestures that complement their speech. Their hands are clearly visible, independent, and unobstructed. Their facial expressions are expressive and full of emotion, enhancing the delivery. The camera remains steady, capturing sharp, clear movements and a focused, engaging presence."
 ]
 
-def slider_value_change(image_path, audio_path, text, num_steps, adaptive_text):
+def slider_value_change(image_path, audio_path, text, num_steps, session_state, adaptive_text):
 
     if adaptive_text:
 
@@ -568,15 +568,15 @@ def slider_value_change(image_path, audio_path, text, num_steps, adaptive_text):
         else:
             text = ADAPTIVE_PROMPT_TEMPLATES[2]
             
-    return update_generate_button(image_path, audio_path, text, num_steps), text
+    return update_generate_button(image_path, audio_path, text, num_steps, session_state), text
 
     
-def update_generate_button(image_path, audio_path, text, num_steps):
+def update_generate_button(image_path, audio_path, text, num_steps, session_state):
 
     if image_path is None or audio_path is None:
         return gr.update(value="⌚ Zero GPU Required: --")
 
-    duration_s = get_duration(image_path, audio_path, text, num_steps, None, None)
+    duration_s = get_duration(image_path, audio_path, text, num_steps, session_state, None)
     duration_m = duration_s / 60
     
     return gr.update(value=f"⌚ Zero GPU Required: ~{duration_s}.0s ({duration_m:.1f} mins)")
@@ -930,9 +930,9 @@ with gr.Blocks(css=css) as demo:
     )
 
     image_input.upload(fn=preprocess_img, inputs=[image_input, session_state], outputs=[image_input])
-    image_input.change(fn=update_generate_button, inputs=[image_input, audio_input, text_input, num_steps], outputs=[time_required])
-    audio_input.change(fn=update_generate_button, inputs=[image_input, audio_input, text_input, num_steps], outputs=[time_required])
-    num_steps.change(fn=slider_value_change, inputs=[image_input, audio_input, text_input, num_steps, adaptive_text], outputs=[time_required, text_input])
+    image_input.change(fn=update_generate_button, inputs=[image_input, audio_input, text_input, num_steps, session_state], outputs=[time_required])
+    audio_input.change(fn=update_generate_button, inputs=[image_input, audio_input, text_input, num_steps, session_state], outputs=[time_required])
+    num_steps.change(fn=slider_value_change, inputs=[image_input, audio_input, text_input, num_steps, session_state, adaptive_text], outputs=[time_required, text_input])
     adaptive_text.change(fn=check_box_clicked, inputs=[adaptive_text], outputs=[text_input])
     audio_input.upload(fn=apply, inputs=[audio_input], outputs=[audio_input]
     ).then(