Subscription updates in multiple device language models

• US 9,035,884 B2
• Filed: 03/15/2013
• Issued: 05/19/2015
• Est. Priority Date: 10/17/2012
• Status: Active Grant

• Current Assignee: Microsoft Technology Licensing LLC (Microsoft Corporation)
• Original Assignee: Nuance Communications, Inc. (Microsoft Corporation)
• Inventors: Phillips, Andrew, Kay, David J., Unruh, Erland, Fu, Eric Jun
• Primary Examiner(s): YEN, ERIC L
• Application Number:
  US13/834,887
• Publication Number:
  US 20140108018A1
• Time in Patent Office: 795 Days
• Field of Search: 345/168, 704/9, 341/22, 341/23, 715/810, 715/811
• US Class Current: 345/168
• CPC Class Codes: G06F 16/273 Asynchronous replication or reconciliation
• CPC Class Codes: G06F 3/0237 using prediction or retrieval techniques
• CPC Class Codes: G06F 40/263 Language identification
• CPC Class Codes: G06F 40/40 Processing or translation of natural language natural language analysis G06F40/20; semantic analysis G06F40/30
• CPC Class Codes: G06F 40/58 Use of machine translation, e.g. for multi-lingual retrieval, for server-side translation for client devices or for real-time translation
• CPC Class Codes: G10L 15/183 using context dependencies, e.g. language models
• CPC Class Codes: G10L 15/22 Procedures used during a speech recognition process, e.g. man-machine dialogue

Abstract

Systems and methods for intelligent language models that can be used across multiple devices are provided. Some embodiments provide for a client-server system for integrating change events from each device running a local language processing system into a master language model. The change events can be integrated, not only into the master model, but also into each of the other local language models. As a result, some embodiments enable restoration to new devices as well as synchronization of usage across multiple devices. In addition, real-time messaging can be used on selected messages to ensure that high priority change events are updated quickly across all active devices. Using a subscription model driven by a server infrastructure, utilization logic on the client side can also drive selective language model updates.

Claims

1. A method of updating local language models in a user device, the method comprising:

receiving, by a master language processing system, a selection of one or more subscriptions,

wherein the subscriptions are for a local language model operating on a user device, and

wherein the local language model is being used by a local language processing system to predictively complete text input by a user of the user device;

receiving, by the master language processing system and from the user device, a triggering event associated with a selected subscription;

determining, by the master language processing system and based on the received triggering event, a subscription package to be used to update the local language model; and

sending, from the master language processing system to the user device, the determined subscription package in order to update the local language model to include the subscription package.

2. The method of claim 1, further comprising updating a master language model maintained on a server to include the subscription package.

3. The method of claim 2, further comprising automatically removing the subscription package from the master language model after a removal event.

4. The method of claim 1, wherein the triggering event is based on a location.

5. The method of claim 4, wherein the location is a city, country, point of interest, or GPS location.

6. The method of claim 1, wherein the triggering event is a broadcast signal or a scanned barcode.

7. The method of claim 1, wherein the triggering event is associated with a social group of the user.

8. The method of claim 1, wherein determining the subscription package based on the triggering event includes using crowd sourcing to select the subscription package.

9. The method of claim 1, further comprising charging the user a fee for the selected subscription.

10. At least one machine-readable storage medium, excluding a transitory propagating signal, and containing a set of instructions, which when executed by one or more processors, cause a data processor to:

receive a selection of one or more subscriptions,

wherein the subscriptions are for a local language model operating on a user device, and

wherein the local language model is being used by a local language processing system to predictively complete text input by a user of the user device;

receive, from the user device, a triggering event associated with a selected subscription;

determine, based on the received triggering event, a subscription package to be used to update the local language model; and

send, to the user device, the determined subscription package in order to update the local language model to include the subscription package.

11. The machine-readable storage medium of claim 10, wherein the set of instructions, when executed by the one or more processors, further cause the data processor to update a master language model maintained on a server to include the subscription package.

12. The machine-readable storage medium of claim 11, wherein the set of instructions, when executed by the one or more processors, further cause the data processor to automatically remove the subscription package from the master language model after a removal event.

13. The machine-readable storage medium of claim 10, wherein the triggering event is based on a location.

14. The machine-readable storage medium of claim 13, wherein the location is a city, country, point of interest, or GPS location.

15. The machine-readable storage medium of claim 10, wherein the triggering event is a broadcast signal or a scanned barcode.

16. The machine-readable storage medium of claim 10, wherein the triggering event is associated with a social group of the user.

17. The machine-readable storage medium of claim 10, wherein the set of instructions, when executed by the one or more processors, further cause the data processor to determine the subscription package based on the triggering event includes using crowd sourcing to select the subscription package.

18. The machine-readable storage medium of claim 10, wherein the set of instructions, when executed by the one or more processors, further cause the data processor to charge the user a fee for the selected subscription.

19. A system comprising:

at least one processor and memory;

means for receiving a selection of one or more subscriptions,

wherein the subscriptions are for a local language model operating on a user device, and

wherein the local language model is being used by a local language processing system to predictively complete text input by a user of the user device;

means for receiving, from the user device, a triggering event associated with a selected subscription;

means for determining, based on the received triggering event, a subscription package to be used to update the local language model; and

means for sending, to the user device, the determined subscription package in order to to update the local language model to include the subscription package.

20. The system of claim 19, further comprising means for updating a master language model maintained on a server to include the subscription package.

21. The system of claim 19, wherein the triggering event is based on a location, a broadcast signal, or a scanned barcode.