Accessibility Development Challenge

Improve a PDF-to-Audio Script for Blind Users

We invite developers to improve an English-language script that converts PDF documents into spoken audio. The goal is to make digital documents easier to access for blind and visually impaired people.

View Instructions Review Technical Requirements
Challenge Accessible PDF-to-Audio Reader
Partner Repository Reimbursor PDF Reader
Submission Source Code, Evidence, and Email

Part One

Prepare a Participant Information PDF

Each participant or team must prepare a short PDF containing the information required to identify the submission and evaluate relevant experience.

  1. Create the document

    Prepare a clearly formatted PDF written in English, with a maximum length of three pages.

  2. Include participant details

    • Full name or team name
    • Professional email address

  3. Describe relevant experience

    Summarize your experience with Python, PDF processing, text-to-speech systems, accessibility, or assistive technology.

  4. Explain your motivation

    Briefly explain why you are interested in improving digital accessibility for blind and visually impaired users.

  5. Name the PDF correctly

    Use the following format: ParticipantName_PDFReaderChallenge.pdf

Part Two

Improve the PDF-to-Audio Script

Participants must improve the existing English-language script available in our partner Reimbursor's repository (feel free to take a look at the code anytime). The updated solution should make PDF content easier to listen to, navigate, and understand.

Core Functionality

The script must continue to extract readable text from PDF files and reproduce that content as spoken audio.

Accessibility

The command-line prompts, messages, documentation, and controls must be clear and usable with screen readers.

Audio Quality

Improve pronunciation, pacing, pauses, error handling, or navigation between pages and sections where possible.

Reliability

The program should handle common PDF errors, empty documents, unsupported content, and missing dependencies gracefully.

Documentation

Update the README with clear installation, execution, testing, and troubleshooting instructions written in English.

Code Quality

Keep the solution understandable and maintainable, with meaningful names, focused functions, and useful comments.

Human-Generated Test Files: The improved script must be tested exclusively with PDF files created by human authors. PDF documents generated through artificial intelligence tools must not be used as test evidence for this challenge.

Part Three

Submit Evaluation Evidence

To evaluate each improved script, participants must provide technical evidence showing that the application was executed successfully with eligible human-generated PDF files.

  1. Prepare the evidence

    Include test results, the names of the tested PDF files, and a short explanation of the improvements.

  2. Send the evidence to the evaluation endpoint

    Submit the required evidence to https://news.reimbursor.info/pdf-reader using the format and request method provided by the organizers.

  3. Send the email submission

    Email the participant PDF, source-code repository link, evidence summary, and execution instructions to the address below.

  4. Use the required subject line

    Use: PDF Reader Accessibility Challenge – Participant Name

Required Submission Materials: Include setup instructions, and execution instructions needed by the evaluation team to install, run, test, and validate the improved script.

Evaluation

Evaluation Criteria

Submissions will be reviewed according to accessibility, technical quality, reliability, usability, documentation, and the practical value of the proposed improvements.

Accessibility

Usability for blind and visually impaired people, including compatibility with screen-reader workflows.

Technical Quality

Code organization, stability, maintainability, and appropriate error handling.

Audio Experience

Clarity, pacing, navigation, pronunciation, and overall listening experience.

Evidence Quality

Completeness and clarity of the submitted test evidence using human-created PDF documents.