SacredTexts AI

AI analysis, concept graph and generative sonification of 48 sacred texts from 8 traditions. Born within the PNRR ITSERR research on AI and religious heritage.

Open demo →

A PWA developed within the ITSERR / UNIMORE / FSCIRE (PNRR) research collaboration on artificial intelligence applied to religious heritage, presented with the From Text to Sound contribution at the ITSERR/RESILIENCE Superseminar 2026. It uses Google Gemini to analyse a corpus of 48 sacred texts from 8 traditions (Christian, Jewish, Islamic, Hindu, Buddhist, Sufi, Taoist, Orthodox). It identifies the tradition, extracts key themes and concepts, and finds cross-references between cultures.

A force-directed graph in p5.js visualises conceptual connections; generative sonification in Tone.js renders the data as soundscapes with five presets (contemplative, crystalline, archaic, ethereal, rhythmic). Results export to JSON-LD and RDF with philological provenance. AI analysis requires a personal Gemini key; graph, corpus and sonification run locally.

Features

Corpus of 48 texts from 8 traditionsGemini analysis with personal keyp5.js force-directed graphGenerative sonification, 5 presetsJSON-LD / RDF export with philological provenanceBilingual IT/EN, installable

Tech Stack

JavaScriptp5.jsTone.jsWeb Audio APIGoogle Gemini APIPWA

Collaborations & commissions

ResearchITSERR / UNIMORE / FSCIRE, PNRR

Collaborations & research

If this work is relevant to your research, get in touch to explore possible collaborations.

Let's talk →