Mar 21 2018
Breakthrough artificial intelligence (AI) for digital automation enables enterprises to achieve straight-through processing (STP) from handwritten forms and 80 percent reduction in manual effort
Captricity, a leader in digital automation for enterprise, is pleased to announce Captricity READ, the first AI-powered software that outperforms humans at reading handwriting. Captricity READ's 91 percent accuracy surpasses human quality and exceeds any previous rates achieved by existing text transcription software.
For Captricity's customers in financial services, healthcare, government and non-profit still relying on manual paper workflows, no-human-touch processing is now a reality. Using READ, enterprises can turn the written word into decision-ready data at a fraction of the typical cost. Surpassing humans in accuracy increases STP by over 50 percent, and this data-first, digital automation means enterprises can now achieve dramatically improved processing times to meet modern customer expectations, without the time and expense of re-engineering hundreds of manual processes.
"Surpassing humans with machine reading is a watershed moment for digital automation," said Kuang Chen, founder and CEO of Captricity. "Enterprises have struggled with automation technology due to extremely low accuracy in reading handwriting and poor-quality images. Unlike any previous technology, Captricity READ is the missing link that drives intelligent automation enterprise-wide."
The engine behind Captricity READ is a deep neural network trained using tens of millions of data points -- representing 35 lifetimes of reading experience -- curated from one billion text transcription tasks Captricity has processed in the past five years. Captricity has the world's largest handwriting training dataset, enabling Captricity READ to be successful on Day One of any deployment. Unlike competitive solutions, Captricity can be integrated into businesses with little custom training and business rules.
Available immediately, Captricity READ expands the company's successful product platform with a new data-first approach to digital automation. It's accuracy claims were measured against a publicly-available benchmark of 3,000 real-life enterprise form fields, and the company is backing up its claims by making this dataset available to other organizations looking to exceed Captricity's breakthrough benchmark.