Voice control hub methods and systems

Abstract

A method for application voice access and control includes receiving a handler registration, receiving an utterance, transmitting the utterance to a cloud layer, receiving an intent and an entity from the cloud layer and dispatching the intent and the entity to the handler. A voice control hub system includes a processor and a memory storing instructions that, when executed by the processor, cause the server to receive a handler registration, receive an utterance, transmit the utterance to a cloud layer, receive an intent and an entity from the cloud layer, and dispatch the intent and the entity to the handler. A non-transitory computer readable medium includes program instructions that when executed, cause a computer to receive a handler, receive an utterance, transmit the utterance to a cloud layer, receive an intent and an entity from the cloud layer, and dispatch the intent and the entity to the handler.

1 . A voice control hub computing system for performing a task within an enterprise business software application, comprising
 one or more processors, and   a memory containing instructions that, when executed, cause the voice control hub computing system to:   receive a handler registration request specifying an object handler to respond to voice commands,   receive an utterance of a user of the enterprise business software application,   transmit the utterance of the user to a remote cloud services layer,   convert the utterance of the user to a text string representing speech-to-text output using a custom speech model,   analyze the text string using one or more trained machine learning models to generate an intent and an entity corresponding to the task,   receive the intent and the entity from the remote cloud services layer, wherein the intent is associated with the entity, and   dispatch the intent and the entity to the object handler.   
     
     
         2 . The voice control hub computing system of  claim 1 , including further instructions that, when executed, cause the voice control hub to:
 receive the text string representing speech-to-text output from the remote cloud services layer.   
     
     
         3 . The voice control hub computing system of  claim 1 , including further instructions that, when executed, cause the voice control hub to:
 receive a channel subscription specifying a channel and one or both of (i) the intent type, and (ii) the entity type, and   based on the channel subscription, dispatch the intent and the entity to the channel.   
     
     
         4 . The voice control hub computing system of  claim 1 , wherein the object handler is a dynamically compiled function. 
     
     
         5 . The voice control hub computing system of  claim 1 , wherein the utterance of the user is received in response to a wake word utterance of the user. 
     
     
         6 . The voice control hub computing system of  claim 1 , including further instructions that, when executed, cause the voice control hub to:
 synthesize a speech response to the user responsive to the utterance of the user, and   cause the speech response to be output in an audio speaker of a computing device of the user.   
     
     
         7 . The voice control hub computing system of  claim 1 , including further instructions that, when executed, cause the voice control hub to:
 set a value in a global context visible to the object handler.   
     
     
         8 . The voice control hub computing system of  claim 1 , wherein the voice control hub computing system is packaged as a shared object that the enterprise business software application can access to enable voice functionality in the application. 
     
     
         9 . A computer-implemented method for enabling voice functionality in an enterprise business software application, comprising:
 receiving a handler registration request specifying an object handler to respond to voice commands,   receiving an utterance of a user of the enterprise business software application,   transmitting the utterance of the user to a remote cloud services layer,   converting the utterance of the user to a text string representing speech-to-text output using a custom speech model,   analyzing the text string using one or more trained machine learning models to generate an intent and an entity corresponding to a task,   receiving the intent and the entity from the remote cloud services layer, wherein the intent is associated with the entity, and   dispatching the intent and the entity to the object handler.   
     
     
         10 . The computer-implemented method of  claim 9 , further comprising:
 receiving the text string representing speech-to-text output from the remote cloud services layer.   
     
     
         11 . The computer-implemented method of  claim 9 , further comprising:
 receiving a channel subscription specifying a channel and one or both of (i) the intent type, and (ii) the entity type, and   based on the channel subscription, dispatching the intent and the entity to the channel.   
     
     
         12 . The computer-implemented method of  claim 9 , further comprising:
 compiling the object handler as a dynamic function.   
     
     
         13 . The computer-implemented method of  claim 9 , further comprising:
 receiving utterance of the user in response to the user uttering a wake word.   
     
     
         14 . The computer-implemented method of  claim 9 , further comprising:
 synthesizing a speech response to the user responsive to the utterance of the user, and   causing the speech response to be output in an audio speaker of a computing device of the user.   
     
     
         15 . The computer-implemented method of  claim 14 , wherein synthesizing the speech response to the user responsive to the utterance to the user is part of a multi-turn interaction with the user. 
     
     
         16 . A non-transitory computer readable medium containing program instructions that when executed, cause a computer to:
 receive a handler registration request specifying an object handler to respond to voice commands,   receive an utterance of a user of the enterprise business software application,   transmit the utterance of the user to a remote cloud services layer,   convert the utterance of the user to a text string representing speech-to-text output using a custom speech model,   analyze the text string using one or more trained machine learning models to generate an intent and an entity corresponding to a task,   receive the intent and the entity from the remote cloud services layer, wherein the intent is associated with the entity, and   dispatch the intent and the entity to the object handler.   
     
     
         17 . The non-transitory computer readable medium of  claim 16  containing further program instructions that when executed, cause a computer to:
 receive the text string representing speech-to-text output from the remote cloud services layer. 
 
     
     
         18 . The non-transitory computer readable medium of  claim 16  containing further program instructions that when executed, cause a computer to:
 receive a channel subscription specifying a channel and one or both of (i) the intent type, and (ii) the entity type, and 
 based on the channel subscription, dispatch the intent and the entity to the channel. 
 
     
     
         19 . The non-transitory computer readable medium of  claim 16  containing further program instructions that when executed, cause a computer to:
 synthesize a speech response to the user responsive to the utterance of the user, and 
 cause the speech response to be output in an audio speaker of a computing device of the user. 
 
     
     
         20 . The non-transitory computer readable medium of  claim 16  containing further program instructions that when executed, cause a computer to:
 set a value in a global context visible to the object handler.