Stream Document Ingestion
Process large TXT, CSV and PDF sources incrementally while keeping memory usage controlled.
Textual genomes. Emergent meaning.
TEXTOME converts source material into independently analyzable textual genomes, preserves the statistical structure of those sources, models observed relationships, and uses those relationships for reconstruction and generation.
TEXTOME begins with observed material. It records what actually exists in the source, how often it appears, and how elements relate to one another. The result is a self-contained entity database that functions as a textual genome for analysis, reconstruction and corpus-level generation.
TEXTOME is a utility-driven desktop application designed for local ingestion and analysis, including large-document workflows that stream data rather than loading entire files into memory.
Process large TXT, CSV and PDF sources incrementally while keeping memory usage controlled.
Convert each source into a self-contained SQLite entity with metadata, genes and ordered source chunks.
Measure characters, observed words and other repeated structures directly from the imported corpus.
Build statistical word-pair and word-triple relationships while preserving continuity across streamed chunks.
Analyze multiple textual genomes together to expose shared frequencies, patterns and emergent relationships.
Reconstruct stored material and generate new text from the statistical relationships observed in selected entities.
Choose the ready-to-run Linux binary package or download the complete source project.
Single-file application build for Linux with the TEXTOME runtime bundled for direct deployment.
Download TEXTOME_0.3.1 Linux BinaryFull Python source package including the ingestion engine, entity database logic, batch processing and build configuration.
Download TEXTOME_0.3.1 SourceTEXTOME is a corpus-derived statistical system for converting source material into independently analyzable textual genomes. It is designed around local processing, transparent source-derived relationships and entity databases that can be analyzed individually or combined into larger corpora.
The project is developed under DarkMatter Software and distributed through the Seaverns software archive.