AI OCR (text recognition) services
Updated: October 17, 2019
2019. Amazon launched AI OCR service - Textract
Amazon has launched a new service Textract for its Web Services customers, and it's like optical character recognition on steroids. It more than just extracts text from documents like its name implies - it can actually identify different document formats and their contents so it can process them properly. The product was designed to be able to recognize if it's taking text from tables and forms from documents, including scanned receipts, tax paperwork or inventories. It then generates structured data that doesn't need human input. Textract can process millions of pages in just a few hours, which can lower document processing costs. Plus, customers can use it even though they don't have previous machine learning experience.
2019. Microsoft launched cloud APIs for form and handwriting recognition
Microsoft introduced several new cognitive services on its Azure Machine Learning cloud platform. First, these are gifts for companies dealing with documents, forms and office notes with handwritten text. The Ink Recognizer and Form Recognizer services allow to transform all these paper documents into digital text and data. Conversation Transcription service - transforms phone dialogs into text with each phrase author recognition. Another new service Personalizer allows you to provide personalized recommendations for website or online store visitors basing on behavioral factors. In addition, Microsoft introduced a new visual interface to create machine learning models. Now even marketers can play with ML. You just need to load the database and specify which parameter you want to predict.
2019. Microsoft added table OCR to mobile Excel
Microsoft has added a new feature to the mobile Excel application that allows users to take a picture of the printed table and convert it into an Excel spreadsheet with editing capabilities. While the feature is only available for Android, but it will also come iOS soon. The feature is available only for Office 365 users. Of course, this feature is intended for simple tables. Complicated tables with merged cells are recognized with errors.
2018. Google Compute Engine adds simple machine learning service
Google launched AutoML - a new service on Google Compute Engine that helps developers — including those with no machine learning (ML) expertise - build custom image recognition models. It’s no secret that it’s virtually impossible for businesses to hire machine learning experts and data scientists these days. There is simply too much demand and not enough supply. The new service allow virtually anybody to bring their images, upload them (and import their tags or create them in the app) and then have Google’s systems automatically create a customer machine learning model for them. The whole process, from importing data to tagging it and training the model, is done through a drag and drop interface. We’re not talking about something akin to Microsoft’s Azure ML studio here, though, where you can use a Yahoo Pipes-like interface to build, train and evaluate models.
2017. Box applied AI to content management
Box has just unveiled Skills and the related SDK, Skills Kit. With these new offerings, organizations and developers now have the ability to pull insights from their massive content stores in Box data sets and apply machine learning to release the intrinsic commercial value in that content. Box is previewing three initial Box Skills, using machine learning tools from Google Cloud and Microsoft Azure to solve common business use cases: Those use cases include: Image recognition (detecting individual objects and concepts in image files, capturing text through optical character recognition (OCR), and automatically adding keyword labels to images to easily build metadata on image catalogs), Audio Transcription & Analysis (uses audio files to create and index a text transcript that can be easily searched and manipulated in a variety of use cases), Video Indexing (analyzes video files to provide text transcription, topic detection and indexing, and facial recognition).