Amazon Technologies, Inc.

Microsoft Technology Licensing, LLC

Autography LLC

Barn2door, Inc.

Andrew Smith

Michael Martin George

Rohit Prasad

Robert N Barrett

Thomas J Waters

Karthik Ramakrishnan

Christopher Schindler

Rafal Kuklinski

Erem Boto

William Chan

Eric Feeny

David Auld

James G Droppo

Ariya Rastrow

Bjorn Hoffmeister

Ariel Nicolas Gordon

Malia Morioka Douglas

Zeyad Rajabi

Eli Joshua Fidler

Matthew D Wood

Robin Winter Troy

Gautam Tiwari

Roland Maximilian Rolf Maas

Melanie C B Gens

Oliver Zheng

Jyoti Chhabra

Mark Lawrence

Veerdhawal Pande

Janelle Joy Maiocco

Samuel James San Nicolas

Raja Bose

Ashley B Cockerill

Jeremy MacRae

Ramya Chaganti

Ryan McCrate

Zexiong Yan

Maarten Van Segbroeck

Geoff Schwab

Techniques for enabling access in a multi-assistant speech processing system are described, where a first assistant system may use components of a second assistant system as data processing components. Runtime operational data and user input data related to the first assistant may be kept separate from the processing data and input data related to the second assistant by propagating a first account ID, for user inputs directed to the first assistant, through the processing pipeline, and using a second account for user inputs directed to the second assistant. A mapping between the first account ID and the second account ID may be accessible to a select number of system components. Handoffs between the two assistants are handled in a manner where data related to one assistant is not accessible by the other assistant.

Data processing in a multi-assistant system

A speech-processing system may determine potential endpoints in a user's speech. Such endpoint prediction may include determining a potential endpoint in a stream of audio data, and may additionally including determining an endpoint score representing a likelihood that the potential endpoint represents an end of speech representing a complete user input. When the potential endpoint has been determined, the system may publish a transcript of speech that preceded the potential endpoint, and send it to downstream components. The system may continue to transcribe audio data and determine additional potential endpoints while the downstream components process the transcript. The downstream components may determine whether the transcript is complete; e.g., represents the entirety of the user input. Final endpoint determinations may be made based on the results of the downstream processing including automatic speech recognition, natural language understanding, etc.

Endpointing in speech processing

Systems and methods are described for managing inventory stock for variable weighted and variable priced agricultural products. A system includes: a memory device storing a data structure that relates a plurality of available base unit values and a plurality of available base unit measurement types to each type of agricultural product type; an interface controller configured to provide an interactive interface for a vendor, the interactive interface including an agricultural product type field, a base unit values field, and a base unit measurement type field; and a processor in communication with the interface controller and the memory device. An example method involves: receiving, from a vendor device via the interactive interface, a selection of agricultural product type in the agricultural product type field, and a selection of a base unit value and a base unit measurement type; and storing a marketplace record to the memory device.

System and method for managing inventory stock for variable weighted and variable priced agricultural products

Techniques for ensuring content output to a user conforms to a quality of the user's speech, even when a speechlet or skill ignores the speech's quality, are described. When a system receives speech, the system determines an indicator of the speech's quality (e.g., whispered, shouted, fast, slow, etc.) and persists the indicator in memory. When the system receives output content from a speechlet or skill, the system checks whether the output content is in conformity with the speech quality indicator. If the content conforms to the speech quality indicator, the system may cause the content to be output to the user without further manipulation. But, if the content does not conform to the speech quality indicator, the system may manipulate the content to render it in conformity with the speech quality indicator and output the manipulated content to the user.

Content output management based on speech quality

Systems and methods for generating authentic digital memorabilia are described. A signor may be provided a digital photograph. The signor's signature, written message, or voice message may be received. Biometric authentication or verification may be performed on the signor's handwriting or voice sample through comparison with stored samples. If the verification signifies a high likelihood signor's handwriting or voice sample is authentic, creation of digital memorabilia is performed by embedding signor's signature or written message in a digital photograph and linking the signor's voice message with the photograph. The digital memorabilia is accompanied by a certificate of authenticity and distributed to a customer or displayed on a website.

Method and system for replaying a voice message and displaying a signed digital photograph contemporaneously

Techniques to manage universal file descriptor models for content files are described. An apparatus may comprise a processor circuit and a file descriptor application operative on the processor circuit to manage file descriptors for content files, the file descriptor application arranged to generate a file descriptor for a content file in accordance with a universal file descriptor model, the universal file descriptor model to comprise a file descriptor surface with multiple file descriptor tiles to present corresponding content parts from the content file, with at least one of the file descriptor tiles defining a content part class representing homogeneous content parts from heterogeneous content file types. Other embodiments are described and claimed.

Techniques to manage universal file descriptor models for content files

The present invention extends to methods, systems, and computer program products for orchestrating notifications between identity platforms and relying parties. Embodiments enable identity platforms to ensure that users consistently receive notifications, even when the identity platforms lack knowledge of which relying parties are notification capable and which relying parties are incapable of notification. Embodiments include an identity platform generating a frameset having a first content frame for displaying a notification and a second content frame for displaying a relying party web page. When the relying party is notification capable, the relying party web page includes functionality for removing the frameset established by the frameset and displaying the notification within the context of the relying party web page. When a client renders the frameset, the client retrieves and renders the relying party web page, removing the frameset and displaying the notification as directed by the relying party.

Orchestration of web notifications

Growing community of inventors

Seattle, WA, United States of America

Andrew Smith