Update app.py

app.py

@@ -1,54 +1,19 @@
+
 import gradio as gr
 from llm_loader import load_model
 from processing import process_input
 from transcription_diarization import diarize_audio
 from visualization import create_charts
-from interview import get_interview_instance
 import time
 import re
 import cv2
 import os
-import traceback
 from config import openai_api_key
 
 # Load the model
 llm = load_model(openai_api_key)
 
-def process_message(message, history, speaker_id):
-    try:
-        print(f"Entering process_message with message: {message}, history: {history}, speaker_id: {speaker_id}")
-        interview = get_interview_instance()
-        response = interview.process_message(message)
-        print(f"Response from interview.process_message: {response}")
-        
-        history = history or []
-        history.append((message, response))
-        
-        print(f"Returning history: {history}")
-        print(f"Type of history: {type(history)}")
-        print(f"Content of history: {history}")
-        return history
-    except Exception as e:
-        print(f"Error in process_message: {str(e)}")
-        print(traceback.format_exc())
-        return []  # Return an empty list in case of error
-
-def start_new_interview(general_impression, speaker_id):
-    try:
-        print(f"Entering start_new_interview with general_impression: {general_impression}, speaker_id: {speaker_id}")
-        interview = get_interview_instance()
-        interview.set_general_impression(general_impression)
-        opening_question = interview.start_interview()
-        print(f"Opening question: {opening_question}")
-        
-        result = [(None, opening_question)]
-        print(f"Returning result: {result}")
-        return result
-    except Exception as e:
-        print(f"Error in start_new_interview: {str(e)}")
-        print(traceback.format_exc())
-        return []  # Return an empty list in case of error
-    
+
 def analyze_video(video_path, progress=gr.Progress()):
     start_time = time.time()
     if not video_path:
@@ -67,32 +32,21 @@ def analyze_video(video_path, progress=gr.Progress()):
     charts, explanations, general_impressions = create_charts(results)
 
     progress(1.0, desc="Charts generation complete.")
-    
+
     end_time = time.time()
     execution_time = end_time - start_time
 
-    output_components = []
-    
+    output_components = []  # transcript
+
     output_components.append(f"Completed in {int(execution_time)} seconds.")
     output_components.append(gr.Textbox(value=transcription, label="Transcript", lines=10, visible=True))
-                
+
     for i, (speaker_id, speaker_charts) in enumerate(charts.items(), start=1):
+        print(speaker_id)
         speaker_explanations = explanations[speaker_id]
         speaker_general_impression = general_impressions[speaker_id]
 
-        # Update the hidden textbox with the general impression
-        output_components.append(gr.Textbox(value=speaker_general_impression, visible=False, label=f"General Impression Speaker {i}"))
-
         with gr.Tab(visible=True):
-            with gr.TabItem(label=f'Interactive Interview'):
-                chatbot = gr.Chatbot(visible=True)
-                msg = gr.Textbox(visible=True)
-                clear = gr.Button("Clear", visible=True)
-                start_interview = gr.Button("Start Interview", visible=True)
-                
-                print(f"Setting up chatbot for speaker {i}")
-                print(f"Type of chatbot: {type(chatbot)}")
-                
             with gr.TabItem(label=f'General Impression'):
                 speaker_section1 = [
                     gr.Markdown(f"### {speaker_id}", visible=True),
@@ -120,26 +74,31 @@ def analyze_video(video_path, progress=gr.Progress()):
                     gr.Textbox(value=speaker_explanations.get("personality", ""),
                                label="Personality Disorders Explanation", visible=True, lines=2)
                 ]
-                
-        output_components.extend([chatbot, msg, clear, start_interview])
+
         output_components.extend(speaker_section1)
         output_components.extend(speaker_section2)
         output_components.extend(speaker_section3)
         output_components.extend(speaker_section4)
 
     # Pad with None for any missing speakers
-    while len(output_components) < 40:  
-        output_components.extend([gr.update(visible=False)] * 13)
+    while len(output_components) < 28:
+        output_components.extend([gr.update(visible=False)] * 9)
 
     return output_components
 
+
 def use_example_1():
     return "examples/Scenes.From.A.Marriage.US.mp4"
+
+
 def use_example_2():
     return "examples/Billie Eilish.mp4"
+
+
 def use_example_3():
     return "examples/Elliot Rodger.mp4"
 
+
 def get_middle_frame(video_path):
     cap = cv2.VideoCapture(video_path)
     total_frames = int(cap.get(cv2.CAP_PROP_FRAME_COUNT))
@@ -153,45 +112,31 @@ def get_middle_frame(video_path):
         return preview_path
     return None
 
+
 with gr.Blocks() as iface:
     gr.Markdown("# Multiple Speakers Personality Analyzer")
-    gr.Markdown("This project provides an advanced AI system designed for diagnosing and profiling personality attributes from video content based on a single speaker or multiple speakers in a conversation.")
+    gr.Markdown(
+        "This project provides an advanced AI system designed for diagnosing and profiling personality attributes from video content based on a single speaker or multiple speakers in a conversation.")
 
     with gr.Row():
         video_input = gr.Video(label="Upload Video")
-    
+
     analyze_button = gr.Button("Analyze")
-            
+
     # Create output components
     output_components = []
-    
     # Add transcript output near the top
     execution_box = gr.Textbox(label="Execution Info", value="N/A", lines=1)
     output_components.append(execution_box)
 
     transcript = gr.Textbox(label="Transcript", lines=10, visible=False)
     output_components.append(transcript)
-                
+
     with open('description.txt', 'r') as file:
         description_txt = file.read()
-                
+
     for n in range(3):  # Assuming maximum of 3 speakers
         with gr.Tab(label=f'Speaker {n + 1}', visible=True):
-            with gr.TabItem(label=f'Interactive Interview'):
-                chatbot = gr.Chatbot(visible=False)
-                msg = gr.Textbox(visible=False)
-                clear = gr.Button("Clear", visible=False)
-                start_interview = gr.Button("Start Interview", visible=False)
-    
-                # Update event listeners
-                msg.submit(process_message, 
-                           inputs=[msg, chatbot, gr.State(f'Speaker {n + 1}')], 
-                           outputs=[chatbot])
-                clear.click(lambda: [], None, chatbot, queue=False)
-                start_interview.click(start_new_interview,
-                                      inputs=[gr.Textbox(visible=False, label=f"General Impression Speaker {n+1}"), gr.State(f'Speaker {n + 1}')],
-                                      outputs=[chatbot])
-                            
             with gr.TabItem(label=f'General Impression'):
                 column_components1 = [
                     gr.Markdown(visible=False),
@@ -213,7 +158,6 @@ with gr.Blocks() as iface:
                     gr.Plot(visible=False),
                     gr.Textbox(label="Personality Disorders Explanation")]
 
-        output_components.extend([chatbot, msg, clear, start_interview])
         output_components.extend(column_components1)
         output_components.extend(column_components2)
         output_components.extend(column_components3)
@@ -227,22 +171,22 @@ with gr.Blocks() as iface:
             gr.Image(preview_1, label="Scenes From A Marriage")
             example_video_1 = gr.Video(example_video_1_path, label="Example 1", visible=False)
             use_example_button_1 = gr.Button("Load Example 1")
-            
+
         with gr.Column(scale=1):
             example_video_2_path = "examples/Billie Eilish.mp4"
             preview_2 = get_middle_frame(example_video_2_path)
             gr.Image(preview_2, label="Billie Eilish")
             example_video_2 = gr.Video(example_video_2_path, label="Example 2", visible=False)
             use_example_button_2 = gr.Button("Load Example 2")
-            
+
         with gr.Column(scale=1):
             example_video_3_path = "examples/Elliot Rodger.mp4"
             preview_3 = get_middle_frame(example_video_3_path)
             gr.Image(preview_3, label="Elliot Rodger")
             example_video_3 = gr.Video(example_video_3_path, label="Example 3", visible=False)
             use_example_button_3 = gr.Button("Load Example 3")
-            
-    gr.HTML("<div style='height: 20px;'></div>")        
+
+    gr.HTML("<div style='height: 20px;'></div>")
     gr.Markdown(description_txt)
     gr.HTML("<div style='height: 20px;'></div>")
     gr.Image(value="appendix/AI Personality Detection flow - 1.png", label='Flowchart 1', width=900)
@@ -284,4 +228,4 @@ with gr.Blocks() as iface:
            )
 
 if __name__ == "__main__":
-    iface.launch()
+    iface.launch()