PDF to Markdown - For Use in Obsidian
Budget: $250 – $750 USD
I need to set up an Obsidian vault to capture content for a series of TTRPG campaigns. Obsidian uses markdown (.md) files. With content highly cross-linked.
The source content is in a series of PDFs. Those PDFs are not set up very well to use AI tools to convert (they use a two-page layout so the AI tools get confused about how the text is meant to be ordered). I've tried that route and the results are not good. So I need someone to do this largely manually.
Specifically requirements:
For each source PDF:
a) Go page-by-page and extract the content
b) Build a page-by-page markdown file containing the content
c) Any images in the source files do not need to be extracted, but do need a placeholder notation in the markdown file, something like, [Image - Region Map] or [Image - Broken Sword]
d) All typesetting (headings, bold, italics, etc) in the source files needs to be preserved in standard markdown in the output
e) Any sections that are meant to be contained together but break over pages (usually a table) need to be converted as single content block
I have five source documents for this project, each pdf source is about 25kb in size but some of that is pictures which do not need conversion.
Skills Needed:
1) Strong reading comprehension (English)
2) Familiarity with PDF file structure
3) Familiarity with Markdown language and .md files
4) Familiarity with Obsidian, how it handles information and content
5) Page Layout (desktop publishing, Quark Express, Adobe Design, Affinity Publisher, etc) familiarity would be a plus in terms being able to set up a sensible structure within the .md files
Timeline:
One month, flexible
The source content is in a series of PDFs. Those PDFs are not set up very well to use AI tools to convert (they use a two-page layout so the AI tools get confused about how the text is meant to be ordered). I've tried that route and the results are not good. So I need someone to do this largely manually.
Specifically requirements:
For each source PDF:
a) Go page-by-page and extract the content
b) Build a page-by-page markdown file containing the content
c) Any images in the source files do not need to be extracted, but do need a placeholder notation in the markdown file, something like, [Image - Region Map] or [Image - Broken Sword]
d) All typesetting (headings, bold, italics, etc) in the source files needs to be preserved in standard markdown in the output
e) Any sections that are meant to be contained together but break over pages (usually a table) need to be converted as single content block
I have five source documents for this project, each pdf source is about 25kb in size but some of that is pictures which do not need conversion.
Skills Needed:
1) Strong reading comprehension (English)
2) Familiarity with PDF file structure
3) Familiarity with Markdown language and .md files
4) Familiarity with Obsidian, how it handles information and content
5) Page Layout (desktop publishing, Quark Express, Adobe Design, Affinity Publisher, etc) familiarity would be a plus in terms being able to set up a sensible structure within the .md files
Timeline:
One month, flexible