Update app.py

app.py

@@ -1,4 +1,5 @@
-# imports
+# Updated full code with safe feedback handling
+
 import gradio as gr
 from huggingface_hub import InferenceClient
 import re
@@ -6,7 +7,7 @@ import random
 import whisper
 from pydub import AudioSegment
 
-# uploading and cleaning the knowledge txt file
+# Load and parse the knowledge file
 def load_questions(file_path):
     with open(file_path, 'r') as f:
         data = f.read()
@@ -16,12 +17,13 @@ def load_questions(file_path):
     for block in question_blocks:
         parts = block.split('Possible Answers:')
         question_text = parts[0].strip()
-        questions.append({'question': question_text})
+        answers_text = parts[1].strip()
+        possible_answers = [ans.strip() for ans in re.split(r'\d+\.\s+', answers_text) if ans.strip()]
+        questions.append({'question': question_text, 'answers': possible_answers})
     return questions
 
 all_questions = load_questions('knowledge.txt')
 
-# creating the questions based on each interview
 questions_by_type = {
     'Technical': [q for q in all_questions if any(keyword in q['question'].lower() for keyword in [
         'function', 'linked list', 'url', 'rest', 'graphql', 'garbage', 'cap theorem', 'sql', 'hash table',
@@ -31,36 +33,31 @@ questions_by_type = {
     'Competency-Based Interview': [q for q in all_questions if any(keyword in q['question'].lower() for keyword in [
         "debugging", "learning", "deadlines", "teamwork", "leadership", "mistake", "conflict", "decision"])],
     'Case': [q for q in all_questions if any(keyword in q['question'].lower() for keyword in [
-        "testing", "financial", "automation", "analysis", "regression", "business", "stakeholder"])]
+        "testing", "financial", "automation", "analysis", "regression", "business", "stakeholder"])] 
 }
 
-# models
 client = InferenceClient("HuggingFaceH4/zephyr-7b-beta")
 whisper_model = whisper.load_model("base")
 
-# whisper audio-to-text function
 def transcribe_audio(file_path):
     try:
-        print(f"📂 Processing audio: {file_path}")
+        print(f"\U0001F4C2 Processing audio: {file_path}")
         audio = AudioSegment.from_file(file_path)
         converted_path = "converted.wav"
         audio.export(converted_path, format="wav")
         result = whisper_model.transcribe(converted_path, fp16=False)
         return result["text"]
     except Exception as e:
-        return f"❌ ERROR: {str(e)}"
+        return f"\u274C ERROR: {str(e)}"
 
-# step 1
 def set_type(choice, user_profile):
     user_profile["interview_type"] = choice
     return "Great! What’s your background and what field/role are you aiming for?", user_profile
 
-# step 2
 def save_background(info, user_profile):
     user_profile["field"] = info
     return "Awesome! Type 'start' below to begin your interview.", user_profile
 
-# updated step 3
 def respond(message, chat_history, user_profile):
     message_lower = message.strip().lower()
 
@@ -112,7 +109,6 @@ def respond(message, chat_history, user_profile):
         chat_history.append((message, feedback))
         return chat_history
 
-    # chatbot fallback
     messages = [{"role": "system", "content": f"You are a professional interviewer conducting a {user_profile['interview_type']} interview for a candidate in {user_profile['field']}."}]
     for q, a in chat_history:
         messages.append({"role": "user", "content": q})
@@ -124,27 +120,33 @@ def respond(message, chat_history, user_profile):
     chat_history.append((message, bot_msg))
     return chat_history
 
-# updated feedback using model
 def generate_feedback(user_profile):
+    feedback = []
     questions = user_profile.get('questions', [])
     answers = user_profile.get('user_answers', [])
-    feedback = []
 
-    for i, (question, user_answer) in enumerate(zip(questions, answers)):
-        prompt = (
-            f"You are an expert interviewer. Please evaluate the following response:\n\n"
-            f"Question: {question['question']}\n"
-            f"Candidate's Answer: {user_answer}\n\n"
-            f"Give a brief, constructive evaluation. Say whether it's a strong, okay, or weak answer and what could be improved."
-        )
+    if not questions or not answers:
+        return "⚠️ Feedback unavailable: Make sure you've completed the interview first."
 
-        response = client.text_generation(prompt, max_new_tokens=200)
-        evaluation = response.strip()
-        feedback.append(f"Question {i+1}:\n{evaluation}\n")
+    for i, user_ans in enumerate(answers):
+        if i >= len(questions):
+            feedback.append(f"⚠️ No matching question for answer {i+1}.")
+            continue
+
+        correct_answers = questions[i].get('answers', [])
+        if not correct_answers:
+            feedback.append(f"❌ No expected answers listed for question {i+1}.")
+            continue
+
+        match = any(ans.lower() in user_ans.lower() for ans in correct_answers)
+        if match:
+            fb = f"Question {i+1}: ✅ Good job!"
+        else:
+            fb = f"Question {i+1}: ❌ Missed some key points: {correct_answers[0]}"
+        feedback.append(fb)
 
     return "\n".join(feedback)
 
-# handle audio
 def handle_audio(audio_file, chat_history, user_profile):
     transcribed = transcribe_audio(audio_file)
     if transcribed.startswith("❌"):
@@ -152,7 +154,6 @@ def handle_audio(audio_file, chat_history, user_profile):
         return chat_history
     return respond(transcribed, chat_history, user_profile)
 
-# UI
 with gr.Blocks() as demo:
     user_profile = gr.State({"interview_type": "", "field": "", "interview_in_progress": False})
     chat_history = gr.State([])
@@ -194,4 +195,4 @@ with gr.Blocks() as demo:
     send_btn.click(lambda: "", None, msg, queue=False)
     audio_btn.click(handle_audio, inputs=[audio_input, chat_history, user_profile], outputs=[chatbot], queue=False)
 
-demo.launch()
+    demo.launch()