media-ingest

To ingest video and audio into a knowledge base, the skill automatically applies transcription and text extraction to generate summaries, highlights, and timestamps. It processes multimedia files to create searchable brain pages.

Yes, you can extract text and entities from PDFs and books automatically. The skill performs entity extraction to link mentioned people and companies to existing brain pages, generating detailed summaries and highlights from the ingested content.

The best way to process screenshots and GitHub repos for text extraction is using automated ingestion with OCR and NLP techniques. This skill captures the content, extracts the text, and generates linked brain pages with summaries.

Yes, this content ingestion tool supports entity linking to existing pages. During the multimedia processing phase, it automatically extracts mentioned people and companies and links them directly to your current brain pages.

To ingest a book into your knowledge base, use the command "ingest this book 'Title of Book.pdf'". The skill will automatically process the file, extract text, and generate a brain page with analysis and highlights.