Re:Auntie PDF – an open source app built using Mistral OCR
by 0x62 on Friday March 07, 2025 @11:39PM
(#43297550)
I’d imagine their capabilities mirror that of Mistral OCR [1]. Mistral outputs markdown, the image would have to be convertible to a reasonably useful markdown structure (charts, tables etc).
Re:Auntie PDF – an open source app built using Mistral OCR
by themanmaran on Friday March 07, 2025 @11:48PM
(#43297580)
This highlights the biggest issue I've found with Mistral OCR. Many of the documents I upload are entirely classified as images, which means no OCR is being run.
Pretty much anything with a different colored background gets returned as (image)[image_001].
Example: https://omni-demo-data.s3.us-east-1.amazonaws.com/test/17398...
Re:Auntie PDF – an open source app built using Mistral OCR
by mtillman on Saturday March 08, 2025 @12:19AM
(#43297691)
LLMs tend to be a hammer in search of a nail when it comes to documents that have imagery. We decided on CV models which results in a high 90s midpoint for the docs our customers care about. If you can afford to go with a cv pipeline, it can outperform all of the LLMs by some margin.
Re:Auntie PDF – an open source app built using Mistral OCR
by bilater on Saturday March 08, 2025 @11:04AM
(#43301192)
Yes - unfortunately it seems they don't read images in the pdf.
Re:Auntie PDF – an open source app built using Mistral OCR
by tmpz22 on Friday March 07, 2025 @11:43PM
(#43297567)
Same vein - YouTube most (all?) llm integrations just scrape the transcript. I -think- google's aistudio does more but I'm unsure.
I mean I get it bulk video processing would be crazy expensive, but at least mention you're only analyzing the transcript especially if you're a paid product.
Re:Auntie PDF – an open source app built using Mistral OCR
by wildzzz on Saturday March 08, 2025 @12:27AM
(#43297728)
Whisper does do text to speech but yes, nearly all just read off the subtitles. There's a video by f4mi on YouTube where she tricked the summarizing bots with off-screen captions filled with nonsense.
Then are you actually doing OCR, or are you just extracting embedded text?