Apparatus and methods for managing resources for a system using voice recognition

• US 9,606,767 B2
• Filed: 06/13/2012
• Issued: 03/28/2017
• Est. Priority Date: 06/13/2012
• Status: Active Grant

• Current Assignee: nVoq, Inc.
• Original Assignee: nVoq, Inc.
• Inventors: Corfield, Charles
• Primary Examiner(s): Augustine, Nicholas
• Application Number:
  US13/495,406
• Publication Number:
  US 20130339858A1
• Time in Patent Office: 1,749 Days
• Field of Search: 715/728
• US Class Current: 1/1
• CPC Class Codes: G06F 3/167 Audio in a user interface, e.g. using voice commands for navigating, audio feedback
• CPC Class Codes: G10L 15/30 Distributed recognition, e.g. in client-server systems, for mobile phones or network applications

Abstract

The technology of the present application provides a method and apparatus to manage speech resources. The method includes detecting a change in a speech application that requires the use of different resources. On detection of the change, the method loads the different resources without the user needing to exit the currently executing speech application. The apparatus provides a switch (which could be a physical or virtual switch) that causes a speech recognition system to identify audio as either commands or text.

Claims

1. A method performed on at least one processor for managing speech resources for a plurality of applications, the method comprising the steps of:

providing data from a client workstation regarding a first speech application and a first set of speech resources being used by the first speech application, the first speech application being a natural language speech recognition application, wherein the first set of speech resources comprises at least a first language model and a first user profile used by the first speech application to convert audio to text, and wherein providing the data regarding the first speech application and the first set of speech resources comprises;

initiating the first speech application at the client workstation,

transmitting identification of a client initiating the first speech application at the client workstation, and

delivering the first set of speech resources based on at least one of the first speech applications or the identification of the client;

receiving data from the client workstation indicative of a switch from the first speech application to a second application;

on receiving the data indicative of the switch from the first speech application to the second application, saving a transcription generated by the first speech application using the first set of speech resources;

determining whether the second application being used at the client workstation requires a second set of speech resources, wherein the second set of speech resources comprises at least a second language model and a second user profile used by the second application to convert audio to text, wherein the first language model and the first user profile are different than the second language model and the second user profile; and

loading the second set of speech resources that comprise at least the second language model to facilitate operation of the second application, wherein the client does not log out of the first speech application prior to initiating the second application.

2. The method of claim 1 wherein the second application is a natural language speech recognition application.

3. The method of claim 1 wherein receiving the data from the client workstation indicative of the switch comprises:

determining a voice command input prompt has been initiated at the client workstation,

receiving the voice command to switch application, and

causing the client workstation to execute the voice command.

4. The method of claim 1 wherein receiving the data from the client workstation indicative of the switch comprises receiving an indication that the second application is an active application.

5. The method of claim 4 wherein the active application is indicated by an active window.

6. The method of claim 2 wherein at least the first set of speech resources or the second set of speech resources comprises:

a command word to initiate a web browser application, and

a recognition of a navigational universal resource locator, wherein the web browser navigates to the universal resource locator.

7. The method of claim 2 wherein at least the first set of speech resources or the second set of speech resources comprises:

a command word, and

a text representation to be input to a textual file wherein the command word operates as a shortcut for causing the text representation to be input.

8. The method of claim 2 wherein at least the first set of speech resources or the second set of speech resources comprises a command word, the command word causing the client workstation to execute a macro.

9. The method of claim 2 wherein at least the first set of speech resources or the second set of speech resources comprises a command word, the command word causing the client workstation to execute a script.

10. The method of claim 1 further comprising the steps of:

saving the first set of speech resources prior to loading the second set of speech resources.

11. An apparatus for managing speech resources for a plurality of applications, the apparatus comprising:

a resource manager operationally linked to a client workstation, wherein;

the resource manager is configured to receive data from the client workstation regarding a first speech application and a first set of speech resources used by the first speech application, the first speech application being a natural language speech recognition application, wherein the first set of speech resources comprises a first language model and a first user profile, and wherein to provide the data regarding the first speech application and the first set of speech resources to the resource manager, the client workstation is configured to;

initiate the first speech application at the client workstation,

transmit identification of a client initiating the first speech application at the client workstation, and

deliver the first set of speech resources based on at least one of the first speech applications or the identification of the client;

the resource manager is configured to receive data from the client workstation when a second application is initiated at the client workstation, wherein the second application uses a second set of speech resources different than the first set of speech resources; and

the resource manager is configured to save a transcription generated by the first speech application and to fetch the second set of speech resources from a memory and transmit the second set of speech resources to be loaded at the client workstation to facilitate the execution of the second application, the second set of speech resources comprising at least a second language model and a second user profile wherein at least one of the second language model models and the second user profile are different than at least one of the first language model models and the first user profile contained in the first set of speech resources, wherein the first speech application does not need to be terminated prior to initiation of the second application.