app.py

import gradio as gr
from transformers import pipeline
from langdetect import detect
import docx
from pypdf import PdfReader

# ========================
# Load models
# ========================

# English summarizer
summarizer_en = pipeline("summarization", model="facebook/bart-large-cnn")

# Vietnamese summarizer
summarizer_vi = pipeline("summarization", model="VietAI/vit5-base-vietnamese-summarization")

# Quiz generator
quiz_generator = pipeline("text2text-generation", model="google/flan-t5-base")

# ========================
# Read file
# ========================

def extract_text(file):
    if file.name.endswith(".pdf"):
        reader = PdfReader(file)
        text = ""
        for page in reader.pages:
            if page.extract_text():
                text += page.extract_text() + "\n"
        return text

    elif file.name.endswith(".docx"):
        doc = docx.Document(file)
        return "\n".join([para.text for para in doc.paragraphs])

    elif file.name.endswith(".txt"):
        return file.read().decode("utf-8")

    else:
        return "Unsupported file format"

# ========================
# Main AI function
# ========================

def summarize_and_quiz(file, num_questions):

    text = extract_text(file)

    if len(text) < 200:
        return "Văn bản quá ngắn / Text too short", ""

    # --- Detect language ---
    lang = detect(text)

    # --- Choose summarizer ---
    if lang == "vi":
        summary = summarizer_vi(
            text,
            max_length=200,
            min_length=80,
            do_sample=False
        )[0]["summary_text"]
        
        prompt = f"""
        Tạo {num_questions} câu hỏi trắc nghiệm ngắn gọn dựa vào nội dung sau:

        {summary}

        Mỗi câu có đáp án đúng.
        """

    else:
        summary = summarizer_en(
            text,
            max_length=200,
            min_length=80,
            do_sample=False
        )[0]["summary_text"]

        prompt = f"""
        Create {num_questions} multiple-choice questions based on the following content:

        {summary}

        Each question must include the correct answer.
        """

    quiz = quiz_generator(prompt, max_length=512)[0]["generated_text"]

    return summary, quiz


# ========================
# Gradio Interface
# ========================

interface = gr.Interface(
    fn=summarize_and_quiz,

    inputs=[
        gr.File(label="📄 Upload PDF / DOCX / TXT"),
        gr.Slider(5, 10, value=5, step=1, label="Số câu hỏi Quiz")
    ],

    outputs=[
        gr.Textbox(label="📌 Tóm tắt nội dung", lines=8),
        gr.Textbox(label="📝 Quiz tự động tạo", lines=12)
    ],

    title="📚 AI TÓM TẮT & TẠO QUIZ (VIỆT + ANH)",
    description="""
    AI tự động nhận diện ngôn ngữ và tóm tắt tài liệu (Việt/Anh), sau đó tạo quiz giúp sinh viên ghi nhớ nhanh.
    """
)

interface.launch()