Update app.py

app.py

@@ -2,47 +2,53 @@ import gradio as gr
 from huggingface_hub import InferenceClient
 import re
 import random
-
-# Load questions (your original backend)
-def load_questions(file_path):
-    with open(file_path, 'r') as f:
-        data = f.read()
-    question_blocks = re.split(r'Question:\s*', data)[1:]
-    questions = []
-    for block in question_blocks:
-        parts = block.split('Possible Answers:')
-        question_text = parts[0].strip()
-        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')
-
-# Question categorization (same as your existing code)
-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',
-        'stack', 'queue', 'recursion', 'reverse', 'bfs', 'dfs', 'time complexity', 'binary search tree',
-        'web application', 'chat system', 'load balancing', 'caching', 'normalization', 'acid', 'indexing',
-        'sql injection', 'https', 'xss', 'hash', 'vulnerabilities'])],
-    '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"])] 
-}
-
+import whisper
+from pydub import AudioSegment
+# models
 client = InferenceClient("HuggingFaceH4/zephyr-7b-beta")
-
-# Backend logic (all functions same as before — no changes)
+whisper_model = whisper.load_model("base")
+
+# whisper audio-to-text function
+def transcribe_audio(file_path):
+    try:
+        print(f"📂 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)}"
+# setting up the users profile (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
 
+# generate question using LLM
+def generate_question(user_profile):
+    system_prompt = f"You are a professional interviewer conducting a {user_profile['interview_type']} interview for a candidate in {user_profile['field']}. Generate one thoughtful, clear, and concise interview question."
+    messages = [{"role": "system", "content": system_prompt}]
+    response = client.chat_completion(messages, max_tokens=100, stream=False)
+    return response.choices[0].message.content.strip()
+
+# generate feedback using LLM
+def generate_feedback_llm(user_profile):
+    feedback = []
+    for i, (question, answer) in enumerate(zip(user_profile.get("questions", []), user_profile.get("user_answers", []))):
+        messages = [
+            {"role": "system", "content": f"You are a professional interviewer providing feedback for a candidate's response in a {user_profile['interview_type']} interview for a {user_profile['field']} role."},
+            {"role": "user", "content": f"Question: {question}\nAnswer: {answer}\nPlease give specific, constructive feedback."}
+        ]
+        response = client.chat_completion(messages, max_tokens=150, stream=False)
+        feedback.append(f"Question {i+1}: {response.choices[0].message.content.strip()}")
+    return "\n\n".join(feedback)
+
+# step 3: interview loop
 def respond(message, chat_history, user_profile):
     message_lower = message.strip().lower()
 
@@ -52,74 +58,61 @@ def respond(message, chat_history, user_profile):
         return chat_history
 
     if message_lower == 'start':
-        interview_type = user_profile['interview_type']
-        selected_questions = questions_by_type.get(interview_type, [])
-        random.shuffle(selected_questions)
-        selected_questions = selected_questions[:10]
-
-        user_profile['questions'] = selected_questions
-        user_profile['current_q'] = 0
+        user_profile['questions'] = []
         user_profile['user_answers'] = []
+        user_profile['current_q'] = 0
         user_profile['interview_in_progress'] = True
 
-        intro = f"Welcome to your {interview_type} interview for a {user_profile['field']} position. I will ask you up to 10 questions. Type 'stop' anytime to end."
-        first_q = f"First question: {selected_questions[0]['question']}"
+        intro = f"Welcome to your {user_profile['interview_type']} interview for a {user_profile['field']} position. I will ask you up to 10 questions. Type 'stop' anytime to end."
+        first_q = generate_question(user_profile)
+        user_profile['questions'].append(first_q)
         chat_history.append((message, intro))
-        chat_history.append(("", first_q))
+        chat_history.append(("", f"First question: {first_q}"))
         return chat_history
 
     if message_lower == 'stop' and user_profile.get("interview_in_progress"):
         user_profile['interview_in_progress'] = False
-        bot_msg = "Interview stopped. Type 'feedback' if you'd like me to analyze your answers."
+        bot_msg = "Interview stopped. Type 'feedback' if you'd like me to analyze your answers. Thanks for interviewing with Intervu!"
         chat_history.append((message, bot_msg))
         return chat_history
 
+    if message_lower == 'feedback':
+        feedback = generate_feedback_llm(user_profile)
+        chat_history.append((message, feedback))
+        return chat_history
+
     if user_profile.get("interview_in_progress"):
-        q_index = user_profile['current_q']
         user_profile['user_answers'].append(message)
-        q_index += 1
-        user_profile['current_q'] = q_index
+        user_profile['current_q'] += 1
 
-        if q_index < len(user_profile['questions']):
-            bot_msg = f"Next question: {user_profile['questions'][q_index]['question']}"
+        if user_profile['current_q'] < 10:
+            next_q = generate_question(user_profile)
+            user_profile['questions'].append(next_q)
+            bot_msg = f"Next question: {next_q}"
         else:
             user_profile['interview_in_progress'] = False
-            bot_msg = "Interview complete! Type 'feedback' if you'd like me to analyze your answers."
-        chat_history.append((message, bot_msg))
-        return chat_history
+            bot_msg = "Interview complete! Type 'feedback' if you'd like me to analyze your answers. Thanks for interviewing with Intervu!"
 
-    if message_lower == 'feedback':
-        feedback = generate_feedback(user_profile)
-        chat_history.append((message, feedback))
+        chat_history.append((message, bot_msg))
         return chat_history
 
-    # Normal chatbot conversation
-    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})
-        messages.append({"role": "assistant", "content": a})
-    messages.append({"role": "user", "content": message})
-
+    # fallback LLM response
+    messages = [
+        {"role": "system", "content": f"You are a professional interviewer conducting a {user_profile['interview_type']} interview for a candidate in {user_profile['field']}."},
+        {"role": "user", "content": message}
+    ]
     response = client.chat_completion(messages, max_tokens=150, stream=False)
-    bot_msg = response.choices[0].message.content
+    bot_msg = response.choices[0].message.content.strip()
     chat_history.append((message, bot_msg))
     return chat_history
 
-def generate_feedback(user_profile):
-    feedback = []
-    questions = user_profile.get('questions', [])
-    answers = user_profile.get('user_answers', [])
-    for i, user_ans in enumerate(answers):
-        correct_answers = questions[i]['answers']
-        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 key points: {correct_answers[0]}"
-        feedback.append(fb)
-    return "\n".join(feedback)
-
-
+    # handle audio input
+def handle_audio(audio_file, chat_history, user_profile):
+    transcribed = transcribe_audio(audio_file)
+    if transcribed.startswith("❌"):
+        chat_history.append(("Audio input", transcribed))
+        return chat_history
+    return respond(transcribed, chat_history, user_profile)
 # The new Intervu 2.0 UI with your design!
 with gr.Blocks(css="""
     body { background-color: #161b24; font-family: 'Nato', sans-serif !important; }