Amazon Technologies, Inc.

Vikram Kumar Gundeti

Fred Torok

Frederic Johan Georges Deramat

Peter Paul Henri Carbon

Frédéric Johan Georges Deramat

Rohan Mutagi

Michael Alan Pogue

John Daniel Thimsen

Bo Li

Peter Spalding VanLund

Dharini Sundaram

Brian Alexander Oliver

Christo Frank Devaraj

Tony Roy Hardie

James Marvin Freeman, Ii

Rongzhou Shen

Benjamin F Jones

Gregory Michael Hart

Scott Ian Blanksteen

Michael F Schwartz

Leopold Bushkin

Allan Timothy Lindsay

Yaser Khan

Munir Mahmood

Shemyla Anwar

Alexander Thomas Loeb

Edward Page Foyle

Mohammed Arif

Jerome Alexander Rolia

Andrew Stuart Huntwork

Ajay Gopalakrishnan

Felix Xiaomeng Wu

Shiladitya Roy

Vikas Jain

Nicolas Anton Medhurst Hertl

Vijitha Raji

Noel Evans

Miguel Alberdi Lorenzo

Lev Danielyan

Eswara Jnana Swaroop Bhupathiraju

Andrew Christopher Miles

Mohammed Yasar Arafatha Abdul Rahim

Balraja Subbiah

Manuel Jesus Leon Rivas

Kurt W Piersol

Jeffrey Penrod Adams

James David Meyers

David Robert Thomas

Michael Dale Whiteley

John Charles Howard

Angela Sun

Mohammed Yasar Arafath Abdul Rahim

Nathan G Leupold

Cory Curtis Manthei

Jacob Michael Simpson

Max David Klein

Peter Spalding Vanlund

Kurt Wesley Piersol

Techniques for a natural language processing (NLP) system to implement more than one assistant are described. The NLP system may receive a natural language input corresponding to more than one user command. The NLP system may respond to a first command, of the natural language input, using a TTS voice of a first NLP system assistant. The NLP system may respond to a second command, of the natural language input, using a TTS voice of a second NLP system assistant.

Multi-assistant natural language input processing

Techniques for using validated communications identifiers of a user's communications profile to resolve entries in another user's contact list are described. When a user imports a contact list, the contact list may include multiple entities related to the same person. The system may identify one of the entries in the contact list that corresponds to a validated communications identifier stored in another user's communications profile. The system may identify other validated communications identifiers in the other user's communications profile and cross-reference them against the entries of the contact list. If the system determines the contact list includes entries for the different validated communications identifiers of the other user, the system may consolidate the entries into a single entry associated with the other user.

Contact list reconciliation and permissioning

A speech recognition platform configured to receive an audio signal that includes speech from a user and perform automatic speech recognition (ASR) on the audio signal to identify ASR results. The platform may identify: (i) a domain of a voice command within the speech based on the ASR results and based on context information associated with the speech or the user, and (ii) an intent of the voice command. In response to identifying the intent, the platform may perform a corresponding action, such as streaming audio to the device, setting a reminder for the user, purchasing an item on behalf of the user, making a reservation for the user or launching an application for the user. The speech recognition platform, in combination with the device, may therefore facilitate efficient interactions between the user and a voice-controlled device.

Speech recognition services

A speech recognition platform configured to receive an audio signal that includes speech from a user and perform automatic speech recognition (ASR) on the audio signal to identify ASR results. The platform may identify: (i) a domain of a voice command within the speech based on the ASR results and based on context information associated with the speech or the user, and (ii) an intent of the voice command. In response to identifying the intent, the platform may perform a corresponding action, such as streaming audio to the device, setting a reminder for the user, purchasing an item on behalf of the user, making a reservation for the user or launching an application for the user. In some instances, the speech recognition platform engages in a back-and-forth dialog with the user in order to properly fulfill the user's request.

Managing dialogs on a speech recognition platform

This disclosure describes, in part, context-driven device arbitration techniques to select a speech interface device from multiple speech interface devices to provide a response to a command included in a speech utterance of a user. In some examples, the context-driven arbitration techniques may include executing multiple pipeline instances to analyze audio signals and device metadata received from each of the multiple speech interface devices which detected the speech utterance. A remote speech processing service may execute the multiple pipeline instances and analyze the audio signals and/or metadata, at various stages of the pipeline instances, to determine which speech interface device is to respond to the speech utterance.

Context driven device arbitration

A system that determines that devices are co-located in an acoustic region and selects a single device to which to send incoming notifications for the acoustic region. The system may group devices into separate acoustic regions based on selection data that selects between similar audio data received from multiple devices. The system may select the best device for each acoustic region based on a frequency that the device was selected previously, input/output capabilities of the device, a proximity to a user, or the like. The system may send a notification to a single device in each of the acoustic regions so that a user receives a single notification instead of multiple unsynchronized notifications. The system may also determine that acoustic regions are associated with different locations and select acoustic regions to which to send a notification based on location.

Outputting notifications using device groups

Synchronized output of audio on a group of devices can comprise sending audio data from an audio distribution master device to one or more slave devices in the group. Scores can be assigned to respective audio playback devices, the scores being indicative of a performance level of the respective audio playback devices acting as a master device. The device with the highest score is designated as a candidate master device and one or more remaining devices are designated as a candidate slave(s). A throughput test is conducted with the highest scoring device acting as the candidate master device. The results of the throughput test are used to determine a master device for a group of devices. Latency of the throughput test can be reduced by using a prescribed time period for completion of the throughput test, and/or by selecting a first group configuration to passes the throughput test.

Selection of master device for synchronized audio

A speech-based system includes a local device in a user premises and a network-based control service that directs the local device to perform actions for a user. The control service may specify a first action that is to be performed upon detection by the local device of a stimulus. In some cases, performing the first action may rely on the availability of network communications with the control service or with another service. In these cases, the control service also specifies a second, fallback action that does not rely upon network communications. Upon detecting the stimulus, the local device performs the first action if network communications are available. If network communications are not available, the local device performs the second, fallback action.

Contingent device actions during loss of network connectivity

A system for enabling end user devices to access third party cloud-based resources. For example, the system may include a first party cloud-based resource for converting sound into a format accessible to the third party cloud-based resource, storing and/or maintaining state information related to the an open communication session between the end user device and the third party cloud-based resources, and converting text-based audio announcements into audio that may be output by the end user device. In some cases, the first party cloud-based resource may transmit user responses together with stored state information to the third party cloud-based resources in a manner that the third part cloud-based resources may treat each interaction with the end user as a separate communication session.

Stateless third party interactions

Systems, methods, and devices for adaptably authenticating a user are disclosed. A device captures a user input, and sends data corresponding thereto to a system. The system determines natural language understanding (NLU) results representing the user input. A user authentication component of the system receives the NLU results and determines a skill configured to perform an action responsive to the user input. The user authentication component adaptably performs user authentication based on a user authentication condition associated with the skill. If the user can be authenticated to the satisfaction of the condition, the NLU results data are sent to the skill, along with an indicator representing the user was authenticated by the system.

User authentication as a service

Techniques for a natural language processing (NLP) system to implement more than one assistant are described. The NLP system may receive a natural language input from a device. The NLP system may also receive one or more signals representing one or more assistants to be implemented with respect to the natural language input. The NLP system may intelligently select an assistant to be invoked with respect to the natural language input. Once the assistant is selected, the NLP system may cause content, output to a user, to have characteristics specific to the assistant.

Multi-assistant natural language input processing to determine a voice model for synthesized speech

Growing community of inventors

Bellevue, WA, United States of America

Vikram Kumar Gundeti