OAKLAND, Calif., Oct. 29, 2019 /PRNewswire/ -- Vidado, the only document extraction AI that guarantees 90%+ accurate handwriting digitization, is announcing the immediate availability of the new Vidado READ API. This API gives existing ECM solutions direct access to Vidado's unparalleled text extraction tool, READ. Any existing ECM, RPA or custom system that needs to digitize handwriting or low resolution documents can quickly embed this API to improve straight-through processing by 30% or more.
"Vidado's dataset is what makes them stand out. The billion plus human verified fields give them unprecedented accuracy in the handwriting space, and it also allows them to deliver those results right away. With their new API, users will be able to start getting those accurate results and feeding them to their downstream systems in as little as 30 minutes," said Humphrey Chen, former Head of Rekognition and Amazon Textract.
The release of this API represents a new embedded model for digitization technologies requiring the lowest implementation effort to achieve remarkable results. Through the API, customers can plug Vidado into any existing workflow and enjoy the most accurate data extraction right from their very first API call. Vidado's machine learning models require no customer-specific training and can handle text snippets from day one.
"Vidado's new single-field oriented API introduces an extraordinary à la carte model of automation across workflows. It immediately enables significantly more accurate data extraction while leveraging existing ECM technology investments. As a Vidado partner, I look forward to sharing this new value-add with our customers," said Mike Mahon, CEO and President at Zia Consulting.
In addition to its rapid implementation, the READ API solves many of the other issues that prevent digitization projects from getting off the ground. Since it's entirely cloud-based, it requires minimal IT and hardware resources - users of the API can start processing data in as little as 30 minutes. The READ API does not require users to relinquish control of their data, customers may opt-in for model training inclusion if they want to, however the opt-in process will not impact the accuracy of their results.
"We want to change the effort required to bring machine learning digitization to customers with existing ECM systems. With this new API any ECM system can achieve results that were previously thought impossible," says Vidado CEO Nowell Outlaw. "Machine learning models take time to train, and if you're model training on new data it can take up to a year to get passable results. With this API, a customer is receiving five years worth of model training from day one. Companies with existing capture technology can plug in Vidado's new API to get 90%+ accuracy levels out of the box across all data types - even handwriting. This has the power to turn legacy processes into intelligent automation with the flip of a switch."
The API is available today at vidado.ai/api and you can access it for free with registration. The API launches on Amazon Web Services (AWS) and will release on Google Cloud Platform (GCP) in the near future. An on-premise solution for customers will be available in 2020.
Vidado's cloud-based, artificial intelligence (AI)-driven, intelligent automation platform identifies, digitizes, validates, and enriches data better than OCR technologies or humans, unlocking new ways for insurance, financial services, and pharmaceutical enterprises to operationalize traditional workflows for increased straight-through processing (STP), streamlined workflows, and increased productivity at lower costs.