AI Handwritten Text Recognition (HTR) Tool

Genealogy Assistant AI Handwritten Text Recognition Tool is a free cross-platform application designed to transcribe collections of historical documents into an easier-to-read format. It serves as a front-end to the OpenAI API, allowing you to convert image files into searchable, transcribed PDF’s with the source image attached. 

This tool can transcribe thousands of images in a single batch, without the need for user intervention. It is generally not meant to provide 100% accurate transcriptions, as AI transcription are still not perfect, but it is designed to make large collections of documents more readable for humans. 

From within the application you can modify the prompt, model and parameters, enabling you to fine-tune how your images are processed. You can also enable multi-threading to have the application work on more than one image at a time.

Download for Windows  |  Download for MacOS |  Download for Linux

Key Features:

• Simple Drag and Drop Interface: No need for complicated command-line tools. Just drag your JPEG files directly into the app.

• Efficient Batch Processing: Transcribe multiple documents simultaneously, 

• Advanced AI Models: Utilize any of OpenAI’s reliable models such as o4-mini-high, with the ability to fall back to GPT-4o if needed to ensure maximum accuracy.

• Highly Customizable Settings: Fine-tune the AI’s transcription behaviour to match your specific research needs with adjustable prompts and token limits.

• Instantly Searchable PDFs: Your documents are converted into PDFs that preserve the original image while embedding high-quality, searchable transcriptions.

Getting Started:

First time Configuration

1. Set Your OpenAI API Key: First, paste your OpenAI API key into the secure “OpenAI API Configuration” field. You can obtain your API key directly from OpenAI’s website. Once entered, your key is safely stored for future sessions.

2. Personalize Your Settings (Optional): The default settings work seamlessly for most users, but if you prefer detailed adjustments, simply click “OCR Settings” to explore advanced configuration options.

Transcribing Your Document

1. Add Files:

   • Drag your JPEG images directly into the application’s drop zone, or

   • Click the area to browse and select your files manually.

2. Review and Manage Files: Your chosen files will display in a convenient list. You can remove any files you no longer want or clear the entire selection at once.

3. Initiate Transcription: Click “Process Files” to begin transcription. A clear progress indicator shows real-time status, and you can access detailed logs or cancel anytime.

4. Access Your Results: Upon completion, a summary of successfully transcribed files appears. Simply open the designated folder to access your fully searchable PDFs.

Advanced Settings

We offer two-tiered accuracy assurance:

• Primary (o4-mini-high): Our standard model balances efficiency and accuracy, customizable to your preferences.

• Fallback (GPT-4o): Automatically engages if the primary model encounters issues, guaranteeing dependable results.

Adjustable Performance Settings

• Concurrent Threads:

  • Single thread: Safest, processing one file at a time.

  • Two to Three threads: Recommended balance for speed and reliability.

  • Four or more threads: Faster but may risk hitting OpenAI’s API rate limits.

Tips and Troubleshooting

• Quality Counts: High-quality images greatly improve transcription accuracy. Ensure clarity and proper lighting for optimal results.

• Check Your Credits: Always verify your OpenAI API account balance and validity of your API key to avoid interruptions.

• Start with Small Batches: Initially transcribe a few documents to confirm settings and accuracy before tackling larger batches.

Report a Bug or Suggest a Feature

If you are experiencing an issue with the application or you have a feature to suggest, use our request form.