Working on Wordspike.com which I launched earlier this month. It exists to help you extract information from videos so that you save time and get a text to work with. It’s freemium model.
I input text and preferably I output JSON but doesn’t matter much as long as it’s somewhat structured.
Ultimately I’d like to extract information like date ranges, specific indications of tool usages (e.g. I have a bunch of data apis with their own individual data and semantic meaning which need to be picked and then a combination of tools to transform the data)
I am creating something along these lines, https://github.com/zero-day-ai, it's meant for security testing, but probably has most of the functionality you need (and you can write plugins fairly easily if not); you can create a prompt repository, defined by a schema that are organized my domains (again, security testing domains, but they can be expanded). If you have any features you'd like to see, or have an ideal workflow feel free to ping me: anthony@zero-day.ai
I'll update UI for the table of contents in a moment.
As for LLMs, my vision for Wordspike’s not to replace human voice, but to act as a filter. It's an add-on to see what video content is valuable to watch in full. It's my attempt to built a counter-weight to the endless amount of shorts and slop I see online.
I built this because I have a love-hate relationship with YouTube. It's an incredible resource, but the platform's design makes it a major productivity trap. I wanted a way to get the knowledge without the time sink.
Wordspike takes any YouTube link and turns it into a clean, readable article with summaries and key insights. It also has a one-click "Send to Kindle" feature. I built the first version in about two weeks using React, Bun, Postgres, and Redis.