An invoice is a document sent to a buyer that specifies the amount and cost of products or services provided by a seller. It indicates the amount and payment terms that must be paid by the buyer.
The Vantage Invoice Skill is delivered as a Process Skill, which consists of five Document Skills handling different regional requirements. The Invoice Skill is trained by ABBYY on a very large document training set of invoices from around the world. The skills are capable of processing invoice documents in a variety of input formats and languages, classifying them based on region, and extracting key data points including header, footer, and line item details with unsurpassed accuracy. The individual regional Invoice Skills support regional tax settings, currencies, and business rules. Extracted data can then be passed on to other business systems and processes with or without manual review and verification.
- Receives and processes invoices in a variety of input formats - scanned images, PDFs, email attachments etc. Provides exceptional accuracy of OCR conversion and data extraction.
- Includes six individual Invoice Document Skills with regional specifics for USA, Canada, Australia & New Zealand, Europe and Spain.
- Automatically detects the invoice region and applies classification, rounting the document to the corresponding Invoice Dokument Skill.
- Pre-trained model automatically detects key fields on the invoice and extracts relevant data (invoice date and number, vendor, PO, amounts, sales taxes, line items, etc.). For a full list of extracted fields, please see the skill documentation.
- Detects Business Unit and Vendor automatically, or allows to match against an existing database for optimal accuracy.
- Includes tax schemes and business rules specific to the individual regions. Performs automatic validations such as required fields, tax amount calculation etc. Further business rules can be added as individually required.
- Can be augmented by human-in-the-loop review and verification based on accuracy thresholds, missing or incorrect information. The system learns from human interaction and constantly improves the model over time.
- Allows further training of the model based on customer documents in real time. Over time significantly improves straight-through-processing rates and reduces the need for human interaction.