Skip to content

OpenPecha/image-transcription

Repository files navigation

TextAlign - Image-Text Alignment Tool

A React-based application for manual text transcription correction with a 3-tier QA pipeline.

Features

  • 3-Tier Quality Assurance Pipeline

    • Transcriber: Receives noisy text and performs initial correction
    • Reviewer: Validates corrections, can approve or reject
    • Final Reviewer: Performs ultimate sanity check for "Gold Standard" approval
  • Role-Based Dashboards: Dedicated views for each user type showing pending tasks, rejected items, and completion metrics

  • Editor Workspace: Side-by-side comparison of source image (with pan/zoom) and editable text field

  • State Machine Logic: Strict tracking of task assignments and rejections with full audit history

Tech Stack

  • Framework: Vite + React 18 + TypeScript
  • Styling: Tailwind CSS + shadcn/ui components
  • Routing: React Router v6
  • State Management: Zustand (UI/auth) + TanStack Query (server state)
  • Image Viewer: react-zoom-pan-pinch

Getting Started

Prerequisites

  • Node.js 18+
  • npm or yarn

Installation

# Install dependencies
npm install

# Start development server
npm run dev

# Build for production
npm run build
span

Task State Machine

Pending → InProgress → AwaitingReview → InReview → AwaitingFinalReview → FinalReview → Completed
                                            ↓                                 ↓
                                        Rejected ←─────────────────────────────┘
                                            ↓
                                        InProgress (re-assigned)

Keyboard Shortcuts

  • Ctrl/Cmd + +: Zoom in image
  • Ctrl/Cmd + -: Zoom out image
  • Ctrl/Cmd + 0: Reset zoom
  • Ctrl/Cmd + S: Save draft (in editor)

License

MIT

About

No description, website, or topics provided.

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages