Noah Siva

Analyzing Voynich Manuscript Using AI Tokenization

The Voynich Manuscript is an undeciphered text from the early 15th century, containing about 170,000 characters and 8,000 words. This talk explores AI tokenization and embeddings, and how they can be applied to unknown languages and scripts. We’ll examine the manuscript’s history, compare my approach with previous research, and explore interesting patterns I found. This could help analyze and preserve lost languages.

About Noah Siva. I’m a high school student passionate about coding and AI. I code for my classes and robotics team, but in my free time, I enjoy working on projects like this one; using AI’s powerful technical capabilities to tackle seemingly obscure and unconventional problems.