Ephesoft, Inc., the leading developer of supervised machine learning-based document capture and analytics solutions, today announces at the Mortgage Bankers Association Technology Solutions Conference & Expo in Detroit the immediate availability of Ephesoft Transact for Mortgage.
The new Ephesoft Transact for Mortgage solution is a cloud-based modular platform pre-trained to recognize and classify more than 600 mortgage document types, dramatically accelerating processing speeds for mortgage lenders who opt to bundle the Ephesoft Transact platform with this tailored classification solution.
“We are continuously developing innovative, cost-effective solutions that solve real problems for our customers and with hundreds of different document types to classify, the mortgage processing industry is ideally suited for document capture innovation and automation,” said Ike Kavas, founder and CEO of Ephesoft, Inc. “Our many mortgage lender and financial services customers will greatly benefit from the improved speeds provided by the Transact for Mortgage platform, which will make it even easier for new and existing customers in the banking and finance fields to benefit from our robust document processing solution, best practices and cloud-based supervised machine learning technology.”
Ephesoft will demonstrate its Transact for Mortgage solution, featuring a new mortgage batch class, at the Technology Showcase today from 10:30 to 11:30 a.m., as well as from its booth #435 throughout the show, being held at the Detroit Marriott at the Renaissance Center.
As the first vertical-specific product from Ephesoft, Transact for Mortgage enables underwriters and mortgage loan processors to upload loan documents into a batch class, which automatically and accurately classifies and separates documents, before validating and exporting them into loan origination systems like Ellie Mae and Mortgage Cadence, along with others. The system automatically recognizes where each document starts and stops within large multi-file PDFs, splitting up individual document types. This new out-of-the-box mortgage software eliminates the hefty professional services fees or internal IT investment required to manually categorize appraisals, lease agreements, tax returns and hundreds of other document types found in mortgage applications.
Large and mid-sized banks, lenders and other mortgage processing organizations using Ephesoft Transact for Mortgage will see configuration and deployment times for an entire enterprise reduced by an estimated 80 percent, from several months to just weeks. Lenders, loan officers and mortgage processors using Transact for Mortgage will enjoy improved accuracy and efficiency to close more loans within days, instead of weeks or months.
Ephesoft Transact for Mortgage provides unprecedented workflow efficiency by automatically determining which documents are needed to process loans and which are ancillary, such as the cover pages, blank sheets of paper and invoices often attached to appraisal documents. Ephesoft does the heavy lifting, removing extraneous content and focusing on high-value documents that can be categorized into batches such as lease agreements, tax returns, loan applications and disclosures. Transact for Mortgage also primes relevant documents for data extraction and business insight providing customers with the option to apply Ephesoft’s patented supervised machine learning for broader use. For more information about Ephesoft’s scalable Software as a Service (SaaS) platform and Transact for Mortgage, visit www.ephesoft.com/solutions/mortgage/.
“Our customers have used Ephesoft’s technology to improve their mortgage document processes,” said Jane Christie, COO of eDocument Solutions, LLC. “Customers have reported accuracy reports of 90 percent or higher, loan processing time reduction of 92 percent and savings of over $100 per loan. The combination of accuracy, consistency and speed for mortgage documents impacts their bottom line and improves customer satisfaction and retention rates.”
Contributed | what does this mean?