Update app.py

app.py

@@ -3,25 +3,14 @@ from transformers import pipeline
 import re
 
 # Initialize pipelines
-# Summarization pipeline with FLAN-T5
 summarizer = pipeline("text2text-generation", model="google/flan-t5-small", tokenizer="google/flan-t5-small")
-
-# Sentiment analysis pipeline
 sentiment_analyzer = pipeline("sentiment-analysis", model="cardiffnlp/twitter-roberta-base-sentiment-latest")
-
-# Automatic speech recognition pipeline for audio
 asr_pipeline = pipeline("automatic-speech-recognition", model="openai/whisper-small")
 
 def convert_to_json(transcript_text):
-    """
-    Convert the transcript into a structured JSON format.
-    Attempts to identify speaker turns based on lines starting with 'Therapist:' or 'Client:'.
-    If no clear pattern is found, the entire transcript is considered one turn.
-    """
     lines = transcript_text.strip().split("\n")
     session_data = []
     
-    # Regex patterns to identify lines with a speaker
     therapist_pattern = re.compile(r"^\s*(Therapist|T):", re.IGNORECASE)
     client_pattern = re.compile(r"^\s*(Client|C):", re.IGNORECASE)
     
@@ -31,61 +20,47 @@ def convert_to_json(transcript_text):
     for line in lines:
         line = line.strip()
         if therapist_pattern.match(line):
-            # If we have accumulated text from previous speaker, store it
             if current_speaker and current_text:
                 session_data.append({"speaker": current_speaker, "text": " ".join(current_text).strip()})
                 current_text = []
-            
             current_speaker = "Therapist"
-            # Remove the speaker prefix
             text_part = therapist_pattern.sub("", line).strip()
             current_text.append(text_part)
-        
         elif client_pattern.match(line):
             if current_speaker and current_text:
                 session_data.append({"speaker": current_speaker, "text": " ".join(current_text).strip()})
                 current_text = []
-            
             current_speaker = "Client"
             text_part = client_pattern.sub("", line).strip()
             current_text.append(text_part)
-        
         else:
-            # Just text, append to current speaker's segment if identified
             if current_speaker is None:
-                # No speaker identified yet, assume unknown
                 current_speaker = "Unknown"
             current_text.append(line)
 
-    # Append the last collected segment
     if current_speaker and current_text:
         session_data.append({"speaker": current_speaker, "text": " ".join(current_text).strip()})
 
-    # If no speakers identified at all and just one big chunk, still return it as JSON
     if not session_data:
         session_data = [{"speaker": "Unknown", "text": transcript_text.strip()}]
 
-    # Create a final JSON structure
     json_data = {"session": session_data}
     return json_data
 
 def analyze_session(transcript, custom_instruction, audio):
-    # If audio is provided, we transcribe it and ignore the text transcript field
+    # If an audio file is provided, transcribe it
     if audio is not None:
-        # Transcribe audio
+        # 'audio' will be the file path if type="filepath"
         asr_result = asr_pipeline(audio)
         transcript_text = asr_result['text']
     else:
-        # Use the provided transcript text
         transcript_text = transcript
-    
+
     if not transcript_text.strip():
         return "Please provide a transcript or an audio file."
-    
-    # Convert transcript to JSON
+
     json_data = convert_to_json(transcript_text)
 
-    # Prepare the prompt for summarization
     prompt = (
         "You are a helpful assistant that summarizes psychotherapy sessions. "
         "The session is provided in JSON format with speaker turns. "
@@ -95,25 +70,20 @@ def analyze_session(transcript, custom_instruction, audio):
         prompt += f" Additionally, {custom_instruction.strip()}"
     prompt += "\n\nJSON data:\n" + str(json_data)
 
-    # Summarize using the LLM
     summary_output = summarizer(prompt, max_length=200, do_sample=False)
     summary = summary_output[0]['generated_text'].strip()
-    
-    # Sentiment analysis of the entire transcript
+
     sentiment_results = sentiment_analyzer(transcript_text)
     main_sentiment = sentiment_results[0]['label']
-    
-    # Simple keyword-based recurring concerns
+
     words = transcript_text.lower().split()
     keywords_of_interest = ["anxiety", "depression", "relationship", "stress", "fear", "goals", "progress", "cognitive", "behavior"]
-    recurring_concerns = [word for word in words if word in keywords_of_interest]
-    recurring_concerns = list(set(recurring_concerns))
+    recurring_concerns = list(set([word for word in words if word in keywords_of_interest]))
     if not recurring_concerns:
         recurring_concerns_str = "No specific recurring concerns identified from the predefined list."
     else:
         recurring_concerns_str = "Recurring concerns include: " + ", ".join(recurring_concerns)
-    
-    # Suggest follow-up topics based on summary
+
     follow_up_suggestions = []
     if "progress" in summary.lower():
         follow_up_suggestions.append("Explore client's perception of progress in more detail.")
@@ -122,12 +92,10 @@ def analyze_session(transcript, custom_instruction, audio):
     if not follow_up_suggestions:
         follow_up_suggestions.append("Consider following up on the emotional themes identified in the summary.")
     follow_up_suggestions_str = " ".join(follow_up_suggestions)
-    
+
     final_output = f"**Summary of Session:**\n{summary}\n\n**Overall Sentiment:** {main_sentiment}\n\n**{recurring_concerns_str}**\n\n**Suggested Follow-Up Topics:** {follow_up_suggestions_str}"
-    
     return final_output
 
-# Build Gradio UI
 description = """# Psychotherapy Session Summarizer
 
 This tool summarizes psychotherapy session transcripts (text or audio) into key themes, emotional shifts, and patterns.
@@ -142,7 +110,7 @@ with gr.Blocks() as demo:
     gr.Markdown(description)
     with gr.Row():
         transcript_input = gr.Textbox(label="Session Transcript (Text)", lines=10, placeholder="Paste the session transcript here...")
-        audio_input = gr.Audio(source="upload", type="file", label="Session Audio (Optional)")
+        audio_input = gr.Audio(type="filepath", label="Session Audio (Optional)")
     custom_instruction_input = gr.Textbox(label="Custom Instruction (Optional)", placeholder="e.g., Focus on anxiety and coping strategies.")
     summarize_button = gr.Button("Summarize")
     output_box = gr.Markdown()