I think Shechkov's approach to using high-quality rendering based on document databases is quite interesting, and it's great that they've been able to create their own databases. Regarding the question of whether to use just text or create a document image, I'd like to ask: have you considered the potential risks of using generated documents, even if they're based on real data? It seems like Regula Forensics could be a useful tool for verification, but I'm curious to know more about how others have used their API in practice. Can anyone share their experience with integrating Regula Forensics into their workflow?