Text web pages, articles, and forum threads PDF documents PowerPoint documents (.pptx) Audio files (mp3/wav) YouTube URLs"...
It doesn't actually state video, do you know if it's summarizing the actual YouTube video, or the subtitles?
Whilst they may seem the same, I can see the former being able to be refined over time based on e.g. in scene visual context
Edit: I missed some of the context, comment above is re you-tldr, not kagi which I haven’t tried.
Text web pages, articles, and forum threads PDF documents PowerPoint documents (.pptx) Audio files (mp3/wav) YouTube URLs"...
It doesn't actually state video, do you know if it's summarizing the actual YouTube video, or the subtitles?
Whilst they may seem the same, I can see the former being able to be refined over time based on e.g. in scene visual context