Hi @all!
I am the founder of the AI Flashcard Maker app KardsAI which I launched with gread success here as Top10 Product twice this year.
After each launch I received a lot of questions and DMs concerning the inner-works of the backend - specifically how the app handles different user documents for processing and text parsing.
The answer was always more complicated than I liked it to be since extracting Text and Markdown text depends on the document type that is selected by the user. Existing OCR and text extraction APIs have a lot of limits, restrictions and are overcomplicated. You have to sign up and host them in azure, aws or elsewhere and combine multiple services in each cloud hoster to come up with a viable solution for your product.
Thats why I created the GetTxt.AI API. With this API you can easily get the text or markdown from any document (pptx, docx, pdf and more), image or even audio and video files. Even more you can not only extract the text but also translate and / or summarise it in one Go.
Moreover the API accepts much higher file sizes for audio and video than other services.
The API is basically the Swiss Army Knife for any LLM or AI Application that needs to extract Text for passing into the AI services.
KardsAI already generated more than 1 Million Flashcards based on this service and I hope others find value in it as well making their live as founders and builders easier.
Let me know if you have questions :)
Timo
KardsAI (Mobile App)