| ---
|
| title: DocuMint
|
| emoji: 🔥
|
| colorFrom: yellow
|
| colorTo: blue
|
| sdk: gradio
|
| sdk_version: 4.44.0
|
| app_file: app.py
|
| pinned: true
|
| ---
|
|
|
| # 🔥 DocuMint - AI Document Processing
|
|
|
| Upload any document and let AI help you understand, refine, and extract information!
|
|
|
| ## Features
|
|
|
| - 📝 **Extract Text** - OCR for images, parse PDFs, Word, Excel
|
| - ✨ **Refine Text** - Clean up messy OCR or handwritten text
|
| - 📋 **Summarize** - Get concise summaries of long documents
|
| - 🔑 **Key Information** - Extract important details automatically
|
| - ❓ **Q&A** - Ask questions about your documents
|
| - 🌐 **Translate** - Convert text to English
|
|
|
| ## Supported Formats
|
|
|
| | Type | Formats |
|
| |------|---------|
|
| | Images | JPG, PNG, BMP, GIF, TIFF, WebP |
|
| | Documents | PDF, DOCX, DOC |
|
| | Spreadsheets | XLSX, XLS |
|
| | Text | TXT, CSV, MD, JSON, XML |
|
|
|
| ## Usage
|
|
|
| ### Web Interface
|
| 1. Upload a document
|
| 2. Select a task
|
| 3. Click "Process Document"
|
|
|
| ### Telegram Bot
|
| 1. Open the bot
|
| 2. Send any document
|
| 3. Tell the bot what to do!
|
|
|
| ## Model
|
|
|
| - **Base**: Qwen2-0.5B-Instruct
|
| - **LoRA**: Custom fine-tuned adapters from [himu1780/DocuMint-Models](https://huggingface.co/himu1780/DocuMint-Models)
|
|
|
| ## Links
|
|
|
| - 📦 [DocuMint-Models](https://huggingface.co/himu1780/DocuMint-Models)
|
| - 📊 [DocuMint-Data](https://huggingface.co/datasets/himu1780/DocuMint-Data)
|
| - 🏆 [DocuMint Train](https://huggingface.co/spaces/himu1780/DocuMint-Train)
|
|
|
