Creative Commons Zero (CC0) licenses © jplenio
The fifteenth eArchiving Initiative training webinar will explore how AI technologies can streamline archiving processes, enhance data accessibility, and ensure the longevity and usability of digital archives. The webinar will provide specific examples and use cases that highlight the transformative impact of AI on digital archiving processes, focusing on three separate AI Proof of Concept (PoC) solutions that will be shared with participants as Docker containers:
- Optical Character Recognition (OCR): Learn how to enhance and utilise an AI-based OCR model.
- Automatic Speech Recognition (ASR): Understand how ASR makes spoken language accessible and analysable.
- Text-to-Speech (TTS): Experience the power of TTS in bringing text to life through natural-sounding speech synthesis.
By leveraging AI, it is possible to automate various aspects of the archiving process, making archived content easily searchable and retrievable. AI-driven OCR can enhance the accuracy of text recognition, reducing manual errors and saving time. ASR can convert spoken content into text, making audio archives searchable and analysable, thus preserving valuable spoken information for future use. TTS can transform text into natural-sounding speech, making archived content more accessible to individuals with visual impairments and those who prefer auditory learning.
Additionally, these AI solutions can help maintaining the integrity and authenticity of digital archives, ensuring that the information remains accurate and unaltered over time. They can also assist in migrating data to new formats, safeguarding against technological obsolescence and ensuring that digital archives remain usable and accessible in the long term.
At the end of the webinar, attendees will have gained insights into freely available open AI models and how to incorporate these within organisational workflows. They will also understand why these are beneficial for electronic archiving.
The training webinar will be presented by Dr Anssi Jääskeläinen of Xamk (South-Eastern Finland University of Applied Sciences) on Thursday, 17 April 2025, 10.00–11.00 CEST.
Technical Requirements: Participants wishing to implement the AI solutions should have either a Linux environment with Python 3.1x or the capacity to run Docker containers. A basic understanding of Python programming is also essential for effectively experimenting with the provided AI PoC solutions.
Practical information about the webinar will be provided once registration is complete.