Hugging Face's logo

Spaces:
g0th
/
Studymaker2
Sleeping

App Files Community
Studymaker2
File size: 3,353 Bytes 
f40afe0
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
cdefe42
f40afe0
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
 
import gradio as gr
import google.generativeai as genai
from transformers import pipeline
import json
from tempfile import NamedTemporaryFile
from ppt_parser import transfer_to_structure  # <- FIXED import

# ✅ Use your Gemini API key
GOOGLE_API_KEY = "AIzaSyA8fWpwJE21zxpuN8Fi8Qx9-iwx3d_AZiw"
genai.configure(api_key=GOOGLE_API_KEY)

summarizer = pipeline("summarization", model="facebook/bart-large-cnn")
gemini_model = genai.GenerativeModel("models/gemini-1.5-flash")

extracted_text = ""

def extract_text_from_pptx_json(parsed_json: dict) -> str:
    extracted_text = ""
    for slide_key, slide in parsed_json.items():
        for shape_key, shape in slide.items():
            if shape.get('type') == 'group':
                group = shape.get('group_content', {})
                for _, group_shape in group.items():
                    if group_shape.get('type') == 'text':
                        for para_key, para in group_shape.items():
                            if para_key.startswith("paragraph_"):
                                extracted_text += para.get("text", "") + "\n"
            elif shape.get('type') == 'text':
                for para_key, para in shape.items():
                    if para_key.startswith("paragraph_"):
                        extracted_text += para.get("text", "") + "\n"
    return extracted_text.strip()

def handle_pptx_upload(pptx_file):
    global extracted_text
    with NamedTemporaryFile(delete=False, suffix=".pptx") as tmp:
        tmp.write(pptx_file.read())
        tmp_path = tmp.name

    parsed_json_str, _, _ = transfer_to_structure(tmp_path, "images")
    parsed_json = json.loads(parsed_json_str)
    extracted_text = extract_text_from_pptx_json(parsed_json)
    return extracted_text or "No readable text found in slides."

def summarize_text():
    global extracted_text
    if not extracted_text:
        return "Please upload and extract text from a PPTX file first."
    summary = summarizer(extracted_text, max_length=200, min_length=50, do_sample=False)[0]['summary_text']
    return summary

def clarify_concept(question):
    global extracted_text
    if not extracted_text:
        return "Please upload and extract text from a PPTX file first."
    prompt = f"Context:\n{extracted_text}\n\nQuestion: {question}"
    response = gemini_model.generate_content(prompt)
    return response.text if response else "No response from Gemini."

# ✅ Gradio Interface
with gr.Blocks() as demo:
    gr.Markdown("## 🧠 AI-Powered Study Assistant for PowerPoint Lectures")

    pptx_input = gr.File(label="📂 Upload Lecture File", file_types=[".pptx", ".pdf", ".docx"])
    extract_btn = gr.Button("📜 Extract & Summarize")

    extracted_output = gr.Textbox(label="📄 Extracted Text", lines=10, interactive=False)
    summary_output = gr.Textbox(label="📝 Summary", interactive=False)

    extract_btn.click(handle_pptx_upload, inputs=[pptx_input], outputs=[extracted_output])
    extract_btn.click(summarize_text, outputs=[summary_output])

    question = gr.Textbox(label="❓ Ask a Question")
    ask_btn = gr.Button("💬 Ask Gemini")
    ai_answer = gr.Textbox(label="🤖 Gemini Answer", lines=4)

    ask_btn.click(clarify_concept, inputs=[question], outputs=[ai_answer])

if __name__ == "__main__":
    demo.launch(server_name="0.0.0.0", server_port=7860, share=True)