Amazon releases AI web services

Amazon releases AI web services

Until recently, creating artificial intelligence products was the province of graduate programs at the nation’s top universities and research shops at a select few technology firms. Last month, however, Amazon Web Services, released three AI products that will allow developers to build conversational user experiences for web, mobile and connected-device apps.

With the AI services, developers can build apps that can understand natural language, turn text into life-like speech, have conversations using voice or text, analyze images and recognize faces, objects and scenes. They’re called Amazon Lex, Amazon Polly and Amazon Rekognition.

“The combination of better algorithms and broad access to massive amounts of data and cost-effective computing power provided by the cloud is making AI a reality for application developers,” said Raju Gulabani, AWS' vice president for databases, analytics and AI. For years, the company has been developing AI technologies to predict what customers might like to read, improve efficiency in fulfillment centers and power the virtual assistant, Alexa, he said.  

The three fully managed Amazon AI services will give developers easy-to-use and cost-effective access to the technology.

Lex provides the same automatic speech recognition technology and natural language understanding found in Amazon’s smart-home device, Alexa. With Lex, developers can “quickly and easily build sophisticated, natural language, conversational bots,” according to AWS.

Developers can build and test conversational bots directly from the AWS Management Console by typing in a few sample phrases (e.g., “find a flight,” or “book a flight”) along with instructions for getting the required parameters and asking clarifying questions to complete task. Amazon Lex builds the language model and integrates with the appropriate backend service. Developers can also use pre-built enterprise connectors to answer questions like “what are my top 10 accounts in” by fetching data from such enterprise systems as Salesforce, Microsoft Dynamics, Marketo, Zendesk, QuickBooks and HubSpot.

Lex technology can act as an information bot by retrieving the requested information; an application bot by completing a desired task; or as conduit to a “conversational user experiences” in an Internet of Things system.

NASA used Lex in ROV-E, an educational replica of the Mars Rover, to allow the six-wheeled robot to understand and reply to questions about NASA’s Mars program by integrating with AWS services to connect and scale with various data sources to retrieve NASA’s Mars exploration information.

If Lex is the technology that allows an application to understand communication, then Polly is the technology that allows an application to add its own voice to the conversation. It turns text into speech by adding natural-sounding speech capabilities to existing applications like newsreaders and e-learning platforms.  Polly also can be used to create entirely new categories of speech-enabled products.

Developers can send text to Amazon Polly using the software development kit from within the AWS Management Console, and Polly immediately returns an audio stream that can be played directly or stored in a standard audio file format. The tool boasts 47 voices in 24 languages.

The final service announced by AWS, Rekognition, makes it easier to add image analysis to an application, putting abilities like facial recognition and analysis at an app-maker’s fingertips.

With the help of Rekognition, an app can compare images or search for particular objects within a collection of images. Developers can build an app that determines the likelihood that faces in two images are of the same person, thereby verifying a user against a reference photo in near real-time. It can also search for a face in a collection of millions of images similar to a reference image.

Amazon Polly and Rekognition are currently available in most AWS U.S. regions and will expand to additional regions in the coming months, the company said. The Amazon Lex preview is available to customers now.

About the Author

Matt Leonard is a former reporter for GCN.


  • Records management: Look beyond the NARA mandates

    Pandemic tests electronic records management

    Between the rush enable more virtual collaboration, stalled digitization of archived records and managing records that reside in datasets, records management executives are sorting through new challenges.

  • boy learning at home (Travelpixs/

    Tucson’s community wireless bridges the digital divide

    The city built cell sites at government-owned facilities such as fire departments and libraries that were already connected to Tucson’s existing fiber backbone.

Stay Connected