Efficient document processing is crucial for seamless hiring. Our latest iteration of OCR, powered by ChatGPT, brings unparalleled OCR capabilities to the table, showcasing the true power of AI in streamlining document management.
We understand the need for swift and accurate document processing. That's why we're thrilled for you to take advantage of our enhanced OCR solution, driven by the incredible capabilities of ChatGPT. Built on a vast knowledge base, AI sets new standards in OCR technology.
Here's Perry from the Product Support Team for a quick walkthrough on this feature!
Advantages over traditional OCR
Universal Recognition. With ChatGPT's powerful AI algorithms, we can now optically recognize most uploaded documents with remarkable accuracy.
Accurate Verification. Data from the documents are being extracted for a second layer of verification. The system matches the name of the candidate to the name of the document. It can also extract other data such as ID numbers and expiration dates.
No Training Required. Advanced AI algorithms instantly process and validate the data, providing you with organized and verified information, without any additional training or configuration efforts.
Document Variation Identification. Leveraging the remarkable capability of ChatGPT to distinguish between different document variations, Talkpush's OCR solution can effortlessly handle a wide array of government documents.
Setup
To enable this feature for your company, you can contact your Account Manager. ChatGPT-powered OCR will work on documents in the Documents section (under the Templates tab). These documents will be available on the candidate profile once they have been uploaded by:
Candidates filling out the prescreening questions
Candidates replying to a document request via chat or email
Recruiters from the CRM through a manual upload.
Ensure the campaign’s job location coincides with the document’s country of origin. For example, a campaign should be located in the Philippines to verify the candidates’ SSS documents successfully.
Determining a Document’s Status
After the candidate uploads a document, a badge displaying its status will show in their profile, showing one of 5 possible states:
Processing: The initial status for the document. The system is examining the file.
Match: The verification process was completed and the files matched the requested document.
Not a match: The verification process was completed and the files didn’t match the requested document.
Partial Match: The system can't find the owner's name in the document to compare with the candidate's name because there's no name to match.
Sample scenario: Back side of some IDs where the name is only included on the front side.Error: The verification process did not finish. This state will only be shown after all retries are exhausted.
Document Status Candidate Attribute
For each document, a new candidate attribute will be created, reflecting its current status. You will find it in the Candidate’s profile as [Document Name] status.
These attributes can be used to create Autoflows or filter candidates. To reuse an existing attribute, change the Key of that existing attribute to the format “document_name_status”.
Learn more about adding your own attributes here.
FAQs
Q: What type of documents can be recognized?
Talkpush’s AI-powered OCR solution currently supports PDF, PNG, and JPEG documents.
Q: What determines the status of a document with multiple files?
The document verification status will be calculated based on the results of all its files. If all files linked to the document are Verified, the document status will be Matched. On the other hand, if at least one file has a Not a Match result, the whole document will take the Not a Match status.
Q: Will the document Status change if I add or delete a file?
The automatic verification process will be triggered whenever there is a modification in the files within a document. This means the OCR status may change when new files are uploaded, or existing ones removed.
When you delete files, the automatic process will calculate the new Status based on the remaining files’ results.
When you add a file, the document’s state will transition to Processing as the automatic process will try to verify those new files. Once we have new files' results, we use those values and pre-existing files' results to calculate the new document status.
Whenever an unmatched document receives a new file that is Verified, the non-matching file will be automatically moved to the Unrecognized category. This way, we can accurately assess the state based on the correct documents, preventing the status from remaining unchanged due to an incorrect file.
With Talkpush's enhanced OCR solution, you have the tools to revolutionize your document processing, enhance efficiency, and elevate your hiring processes to new heights. Embrace the power of AI today and unlock the full potential of your recruitment workflows!
🤔 Do you have any questions? We want to help! 💚
You can chat with our Support team by clicking the intercom button on the lower section of your Talkpush screen. We're one click away.
Need more help? 📖