About

SummarAIse turns long videos into useful research notes.

The project watches selected long-form channels and converts each video into a clean summary, transcript, topics and source metadata.

How summaries are generated

  1. Video metadata and transcripts are collected from source videos when available.
  2. Long videos are split into chapters or time-based chunks so the model keeps context local and precise.
  3. Local/controlled LLM workflows produce executive summaries, TL;DR blocks, chapter summaries and optional alpha notes.
  4. WordPress stores the public pages, clean URLs, canonical TL;DR meta and shared topic taxonomy.
Transparency

This is a curated research layer, not a replacement for the original video. Source links remain available on every summary page.

Scroll to Top