Amazon Technologies, Inc.

Rashmi Tonge

Rohin Dabas

Troy Dean Schuring

Nnenna Eleanya Okwara

David Robert Thomas

Kevindra Pal Singh

Adam D Baran

Michael James Montgomery

Rajesh Kumar Pandey

Andres Felipe Borja Jaramillo

Isaac Joseph Madwed

Alberto Milan Gutierrez

Julia Kennedy Nemer

Xu Zhang

Maksym Kolodeznyi

David Thomas

Described herein is a system for enabling a user to perform complex goals using multiple skills/applications of an intelligent assistant device. Skills may register as consumers of an action or providers of an action, and the consumer skills may be configured to invoke provider skills to perform actions. The system receives a request to perform an action from a skill along with some action data. The system validates the action data, selects another skill to perform the action, and forwards the request to the selected skill to perform the action.

Multiple skills processing

Techniques for intelligently selecting a component to execute with respect to a natural language user input are described. A natural language processing (NLP) system may receive first data representing a natural language input. The NLP system may determine first and second scores representing first and second confidences that first and second components are to be invoked to perform actions responsive to the natural language input, respectively. Based on the first and second scores, the NLP system may determine further information is needed to determine which of the first or second component is to be invoked. The NLP system may query a user for the further information. Based on the further information, the NLP system may determine third and fourth scores representing third and fourth confidences that the first and second components are to be invoked to perform actions responsive to the natural language input, respectively. The NLP system may determine the third score is greater than the fourth score and, based thereon, cause the first component to perform an action responsive to the original natural language input.

Natural language input processing

Techniques for transferring control of a system-user dialog session are described. A first speechlet component may interact with a user until the first speechlet component receives user input that the first speechlet component cannot handle. The first speechlet component may output an action representing the user input. A system may determine a second speechlet component configured to execute the action. The system may send the second speechlet component a navigator object that results in the second speechlet component handling the user interaction that the first speechlet component could not handle. Once the second speechlet component is finished processing, the second speechlet component may output an updated navigator object, which causes the first speechlet component to either further interact with a user or cause a current dialog session to be closed. The system may additionally maintain a data structure representing calling speechlet components and called speechlet components associated with the session.

Multi-speechlet response

Growing community of inventors

Bellevue, WA, United States of America

Rashmi Tonge