app.py

# imports
import gradio as gr
from huggingface_hub import InferenceClient
import random
import whisper
from pydub import AudioSegment

# 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}")
        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()

    if not user_profile.get("interview_type") or not user_profile.get("field"):
        bot_msg = "Please finish steps 1 and 2 before starting the interview."
        chat_history.append((message, bot_msg))
        return chat_history

    if message_lower == 'start':
        user_profile['questions'] = []
        user_profile['user_answers'] = []
        user_profile['current_q'] = 0
        user_profile['interview_in_progress'] = True

        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(("", 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. 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"):
        user_profile['user_answers'].append(message)
        user_profile['current_q'] += 1

        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. Thanks for interviewing with Intervu!"

        chat_history.append((message, bot_msg))
        return chat_history

    # 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.strip()
    chat_history.append((message, bot_msg))
    return chat_history

# 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)

# UI
with gr.Blocks() as demo:
    user_profile = gr.State({"interview_type": "", "field": "", "interview_in_progress": False})
    chat_history = gr.State([])

    gr.Markdown("# Welcome to Intervu")
    gr.Image(value="images.JPEG", show_label=False, width=200)

    gr.Markdown("### Step 1: Choose Interview Type")
    with gr.Row():
        with gr.Column():
            btn1 = gr.Button("Technical")
            btn2 = gr.Button("Competency-Based Interview")
            btn3 = gr.Button("Case")
        type_output = gr.Textbox(label="Bot response", interactive=False)

    btn1.click(set_type, inputs=[gr.Textbox(value="Technical", visible=False), user_profile], outputs=[type_output, user_profile])
    btn2.click(set_type, inputs=[gr.Textbox(value="Competency-Based Interview", visible=False), user_profile], outputs=[type_output, user_profile])
    btn3.click(set_type, inputs=[gr.Textbox(value="Case", visible=False), user_profile], outputs=[type_output, user_profile])

    gr.Markdown("### Step 2: Enter Your Background")
    background = gr.Textbox(label="Your background and field/goal")
    background_btn = gr.Button("Submit")
    background_output = gr.Textbox(label="Bot response", interactive=False)

    background_btn.click(save_background, inputs=[background, user_profile], outputs=[background_output, user_profile])

    gr.Markdown("### Step 3: Start Interview")
    chatbot = gr.Chatbot(label="Interview Bot")

    with gr.Row():
        msg = gr.Textbox(label="Your message")
        audio_input = gr.Audio(type="filepath", label="🎧 Upload or Record your answer")

    with gr.Row():
        send_btn = gr.Button("Send Text")
        audio_btn = gr.Button("Send Audio")

    send_btn.click(respond, inputs=[msg, chat_history, user_profile], outputs=[chatbot], queue=False)
    send_btn.click(lambda: "", None, msg, queue=False)
    audio_btn.click(handle_audio, inputs=[audio_input, chat_history, user_profile], outputs=[chatbot], queue=False)


if __name__ == "__main__":
    demo.launch()