Hi,
I need some suggestions on form extractor in UiPath.
I am using an on prime orchestrator and planning to use the du in one of my project. But the limitations are we don’t have any AI center enabled and no cloud orchestrator also. I need to extract arround 70 different fields from a pdf document can I use form extractor here without using any API key.
Hi @sanjay_gowda,
Yes, you can use the Form Extractor without an API key. If your document has a fixed layout/template, the Form Extractor is a good choice.
If some fields are not extracted reliably, you can complement it with the Regex Based Extractor for specific fields.
Thanks
I was able to create template and everything but when I try to run the workflow it is giving me error as
Cannot authenticate with DU services : please provide an API key or make sure orchestrator is connected to the correct instance for the ML skill
If you don’t have Automation Cloud/Document Understanding API access and you also don’t have AI Center available on-prem, then you won’t be able to use the Form Extractor directly. It requires either a DU API key or an on-prem AI Center deployment.
Since you mentioned around 70 fields, I would first check how consistent the PDF layout is.
If the documents follow a fixed format and the field values can be identified using labels or predictable patterns, you can try the Regex Based Extractor. It doesn’t have the same API key requirement.
Another option is to extract the PDF text/OCR output and then use Regex or string logic for the required fields.
For 70 fields this can become quite a bit of configuration and maintenance, so if the document layouts vary a lot, I would suggest checking whether DU/AI Center licensing can be enabled rather than building everything with rules.
Hope this helps.