Choose PowerCred.io for hassle-free document processing. Our smart tool uses the latest AI to read and understand any document, no matter the language or format. It’s quick, accurate, and easy to use, freeing you from the burden of manual work. With PowerCred.io, you get more time for what really matters in your business. Simplify and automate your document tasks with us—let's make every document work for you.
About IDP
PowerCred's Intelligent Document Parser (IDP) sets a new standard in data extraction. It seamlessly handles an array of document formats, languages, and types with ease. But what sets IDP apart is its ability to go beyond pre-defined output formats. With the innovative Bring-Your-Own-Schema feature, users can finely tune output fields to meet their specific requirements. This level of customization ensures that IDP is not just a solution, but the solution for diverse data processing needs. Dive deeper below to discover more about the built-in output formats and witness the true power of IDP.
Executing the API
1. Generate Session ID
Our Get Token API is designed to support multiple sessions of one specific user. Here's a short example:
A user initiates a document upload process with their phone number (9890123456) serving as a unique identifier. During this process, they are expected to upload two types of documents: one for identity verification and another for income verification. However, they discover that they only have their identity document ready and lack the necessary income documentation (like a payslip or bank statement). Consequently, they abandon the upload process midway.
After some time, the user acquires the required income document and decides to resume the upload journey. Upon returning, a new session is initiated for the document upload, resulting in the generation of a new session ID. Despite this, the user's unique identifier (the phone number 9890123456) remains unchanged.
Thus, both session IDs are associated with the same user, based on the unique identifier (the phone number). All data collected during these sessions are aggregated under this singular user ID, ensuring a cohesive view of the user's activities and document submissions.
Our "Get Token" API features an option to include a callback URL, enhancing its versatility and functionality for developers. This addition allows for two primary methods of consuming the output from most of our APIs:
Call GET HTTP API: This approach is useful for continuous monitoring of the processing status. Developers can repeatedly call this GET HTTP API to check on the progress. The API should be polled until the data has been fully processed by our system. Once processing is complete, the API will return a JSON response containing the relevant data.
Callback URL: For a more automated process, developers have the option to specify a callback URL when they initiate the API call. This URL is utilized by our system to automatically send the processed data once it's ready. It's crucial that the specified callback URL is publicly accessible and configured to accept HTTP POST requests, as our system does not support the GET method for callbacks. This method ensures that developers receive the processed data directly and efficiently, without the need for continuous polling.
2. Start Document Processing
Our Intelligent Document Parser is designed to handle documents in any format and language that uses the Latin alphabet, streamlining the document processing experience to its core simplicity. This innovative system eliminates the need for document annotation, sampling, labeling, or any form of preliminary training. Users simply need to select the type of document they're uploading, upload the document itself, and that's it!
Our technology is adept at adapting to format variations and changes effortlessly, ensuring a seamless and efficient processing experience without the necessity for any retraining. It's a straightforward and effective solution that works like magic!
To facilitate the document processing, we have 2 APIs:
Parse Document API: Accessible at Parse Document API, this API requires four parameters:
document_type: Supports 10 distinct document types, with details available here.
(Optional) file_url OR file: Accepts PDF, JPG, or PNG files either as a direct multipart upload or via a valid HTTP(S) URL.
(Optional) schema_file: Allows for a custom schema or output format selection for data extraction. Default schema files are listed here. The schema can be uploaded together with the document.
(Optional) document_type: Specifies the document type from our list of 10. If unspecified, returns all documents associated with the session ID.
(Optional) return_json: When set to true, returns the extracted document data in JSON format. By default (false), it provides excel_url (a link to download the data in Excel format) and input_url (a link to download the original uploaded file).