De Codes van het Gregoriaans

Home » Algemeen » Perplexed: AI-supported analysis of signal tones

Perplexed: AI-supported analysis of signal tones

Onbekend's avatar

Vertalen

Follow De Codes van het Gregoriaans on WordPress.com

Archives

A mail, I sent to some of my friends this morning:

Dear ,

Printed books,

Traveling by rail,

Google,

AI

All labeled as The End of Intellectuality or similar.

You know that I have been experimenting with neume-text retrieval. This morning, I tested some recent ideas. The first results are very promising, if not to say that they might be a major breakthrough for the intermedial analysis of signal tones.

In Perplexity Pro, I uploaded 2 scans:

Scan 1: a list with scans of 8 ornamental neumes and their names next to them, copied from U406

Scan 2: a folio from U 406

Command to software: “trace symbol 5, (quilisma) from scan 1 in scan 2 and mention the corresponding complete word to which this symbol applies.”

I could not believe my eyes…. And I even did not ask for their location on the page….

This is only the beginning.

See attachment. 

The ms U 406 is problematic because it is bound far too tight. But, for instance, the two main Hildegard sources I am currently analyzing for their signal tones have been scanned with flat pages and their ‘readability’ is 100%.


Plaats een reactie

Deze site gebruikt Akismet om spam te bestrijden. Ontdek hoe de data van je reactie verwerkt wordt.