Access OCR Index

Hello -
Using OC daily, I want to deploy a more advanced index.
The instance is running on a rather standard web hosting. I don’t have access to the console neither to the server options.
It is therefore impossible to install tesseract and OCR on the server.
My idea: do the processing stuff in advance, before uploading the file.
How can I make this previously indexed file searchable? Are there any ways to exploit this index?
—visoris