System and method for dynamically generating a recognition grammar in an integrated voice navigation services environment

• US 8,326,627 B2
• Filed: 12/30/2011
• Issued: 12/04/2012
• Est. Priority Date: 12/11/2007
• Status: Active Grant

• Current Assignee: Cerence Inc.
• Original Assignee: VoiceBox Technologies, Inc. (Microsoft Corporation)
• Inventors: Kennewick, Michael R., Menaker, Sam, Zimmerman, Bernie, Dichristo, Philippe, Armstrong, Lynn, Guttigoli, Sheetal, Tjalve, Michael, Salomon, Ari, Baldwin, Larry, Cheung, Catherine
• Primary Examiner(s): Vo, Huyen X.
• Application Number:
  US13/341,479
• Publication Number:
  US 20120101809A1
• Time in Patent Office: 340 Days
• Field of Search: 704/247, 704/251, 704/255, 704/257, 704 1- 10, 704/231, 704/235, 704/270, 704/270.1
• US Class Current: 704/257
• CPC Class Codes: G01C 21/3608 using speech input, e.g. using speech recognition
• CPC Class Codes: G06Q 30/0261 based on user location
• CPC Class Codes: G10L 15/00 Speech recognition G10L17/00 takes precedence
• CPC Class Codes: G10L 15/04 Segmentation; Word boundary detection
• CPC Class Codes: G10L 15/08 Speech classification or search
• CPC Class Codes: G10L 15/19 Grammatical context, e.g. disambiguation of the recognition hypotheses based on word sequence rules
• CPC Class Codes: G10L 15/193 Formal grammars, e.g. finite state automata, context free grammars or word networks
• CPC Class Codes: G10L 15/22 Procedures used during a speech recognition process, e.g. man-machine dialogue

Abstract

The system and method described herein may dynamically generate a recognition grammar associated with a conversational voice user interface in an integrated voice navigation services environment. In particular, in response to receiving a natural language utterance that relates to a navigation context at the voice user interface, a conversational language processor may generate a dynamic recognition grammar that organizes grammar information based on one or more topological domains. For example, the one or more topological domains may be determined based on a current location associated with a navigation device, whereby a speech recognition engine may use the grammar information organized in the dynamic recognition grammar according to the one or more topological domains to generate one or more interpretations associated with the natural language utterance.

Claims

1. A method for dynamically generating a recognition grammar in an integrated voice navigation services environment, comprising:

receiving a natural language utterance from an input device coupled to a navigation device, wherein the natural language utterance relates to a navigation context;

identifying a current location associated with the navigation device using a location detection system coupled to the navigation device;

generating, at a conversational language processor, a dynamic recognition grammar that organizes grammar information based on one or more topological domains associated with the current location associated with the navigation device; and

determining one or more affinities between a user that spoke the natural language utterance and one or more peers associated with the user, wherein the dynamic recognition grammar further organizes the grammar information according to the one or more determined affinities;

generating, at a speech recognition engine, one or more interpretations associated with the natural language utterance using the dynamic recognition grammar.

2. The method of claim 1, wherein generating the dynamic recognition grammar includes:

recognizing, at a multi-pass speech recognition module associated with the speech recognition engine, one or more command words in the natural language utterance that define a command in the navigation context;

recognizing, at the multi-pass speech recognition module, one or more location words in the natural language utterance that define a state associated with the command in the navigation context;

wherein the dynamic recognition grammar further organizes the grammar information according to the state associated with the command in the navigation context; and

recognizing, at the multi-pass speech recognition module, one or more additional location words in the natural language utterance that define a city within the state associated with the command in the navigation context, wherein the dynamic recognition grammar further organizes the grammar information according to multiple street addresses in the city within the state associated with the command in the navigation context.

3. The method of claim 2, wherein generating the one or more interpretations associated with the natural language utterance using the dynamic recognition grammar includes:

recognizing, at the speech recognition engine, one or more additional command words in the natural language utterance using the grammar information organized according to the multiple street addresses in the city within the state associated with the command in the navigation context, wherein the one or more additional command words define one or more of the multiple street addresses; and

combining the one or more command words and the one or more additional command words recognized in the natural language utterance to generate the one or more interpretations associated with the natural language utterance.

4. The method of claim 1, wherein the one or more topological domains include a state mapped to the current location associated with the navigation device and generating the dynamic recognition grammar includes:

recognizing, at a multi-pass speech recognition module associated with the speech recognition engine, one or more command words in the natural language utterance that define a command in the navigation context; and

recognizing, at the multi-pass speech recognition module, one or more location words in the natural language utterance that define a city within the state mapped to the current location associated with the navigation device, wherein the dynamic recognition grammar further organizes the grammar information according to multiple street addresses in the city within the state mapped to the current location associated with the navigation device.

5. The method of claim 4, wherein generating the one or more interpretations associated with the natural language utterance using the dynamic recognition grammar includes:

recognizing, at the speech recognition engine, one or more additional command words in the natural language utterance using the grammar information organized according to the multiple street addresses in the city within the state mapped to the current location associated with the navigation device, wherein the one or more additional command words define one or more of the multiple street addresses; and

combining the one or more command words and the one or more additional command words recognized in the natural language utterance to generate the one or more interpretations associated with the natural language utterance.

6. The method of claim 1, wherein the dynamic recognition grammar has a size constrained according to memory or resources available to the navigation device via the integrated voice navigation services environment.

7. The method of claim 1, further comprising modifying the grammar information in the dynamic recognition grammar in response to one or more changes to the one or more topological domains associated with the current location associated with the navigation device.

8. The method of claim 1, wherein the dynamic recognition grammar organizes the grammar information according to geographic chunks based on the one or more topological domains associated with the current location associated with the navigation device.

9. The method of claim 8, further comprising dividing the geographic chunks into multiple tiles, wherein the dynamic recognition grammar further organizes the grammar information according to the multiple tiles associated with the divided geographic chunks.

10. The method of claim 9, further comprising dividing one or more of the multiple tiles into multiple subtiles, wherein the dynamic recognition grammar further organizes the grammar information according to the multiple subtiles associated with the divided one or more of the multiple tiles.

11. The method of claim 8, wherein the geographic chunks include physical proximities that relate to a distance from the current location associated with the navigation device, temporal proximities that relate to a travel time from the current location associated with the navigation device, directional proximities that relate to a directional travel vector from the current location associated with the navigation device, or civil organizational proximities that relate to continents, countries, regions, states, cities, localities, neighborhoods, or communities mapped to the current location associated with the navigation device.

12. A system for dynamically generating a recognition grammar in an integrated voice navigation services environment, comprising:

an input device configured to receive a natural language utterance, wherein the natural language utterance relates to a navigation context;

a location detection system configured to identify a current location associated with a navigation device;

a conversational language processor configured to generate a dynamic recognition grammar that organizes grammar information based on one or more topological domains associated with the current location associated with the navigation device and configured to determine one or more affinities between a user that spoke the natural language utterance and one or more peers associated with the user, wherein the dynamic recognition grammar further organizes the grammar information according to the one or more determined affinities; and

a speech recognition engine configured to generate one or more interpretations associated with the natural language utterance using the dynamic recognition grammar.

13. The system of claim 12, further comprising a multi-pass speech recognition module configured to:

recognize one or more command words in the natural language utterance that define a command in the navigation context;

recognize one or more location words in the natural language utterance that define a state associated with the command in the navigation context, wherein the dynamic recognition grammar further organizes the grammar information according to the state associated with the command in the navigation context; and

recognize one or more additional location words in the natural language utterance that define a city within the state associated with the command in the navigation context, wherein the dynamic recognition grammar further organizes the grammar information according to multiple street addresses in the city within the state associated with the command in the navigation context.

14. The system of claim 13, wherein to generate the one or more interpretations associated with the natural language utterance using the dynamic recognition grammar, the speech recognition engine is further configured to:

recognize one or more additional command words in the natural language utterance using the grammar information organized according to the multiple street addresses in the city within the state associated with the command in the navigation context, wherein the one or more additional command words define one or more of the multiple street addresses; and

combine the one or more command words and the one or more additional command words recognized in the natural language utterance to generate the one or more interpretations associated with the natural language utterance.

15. The system of claim 12, wherein the one or more topological domains include a state mapped to the current location associated with the navigation device, and wherein the system further comprises a multi-pass speech recognition module configured to:

recognize one or more command words in the natural language utterance that define a command in the navigation context; and

recognize one or more location words in the natural language utterance that define a city within the state mapped to the current location associated with the navigation device, wherein the dynamic recognition grammar further organizes the grammar information according to multiple street addresses in the city within the state mapped to the current location associated with the navigation device.

16. The system of claim 15, wherein to generate the one or more interpretations associated with the natural language utterance using the dynamic recognition grammar, the speech recognition engine is further configured to:

recognize one or more additional command words in the natural language utterance using the grammar information organized according to the multiple street addresses in the city within the state mapped to the current location associated with the navigation device, wherein the one or more additional command words define one or more of the multiple street addresses; and

combine the one or more command words and the one or more additional command words recognized in the natural language utterance to generate the one or more interpretations associated with the natural language utterance.

17. The system of claim 12, wherein the dynamic recognition grammar has a size constrained according to memory or resources available to the navigation device via the integrated voice navigation services environment.

18. The system of claim 12, wherein the conversational language processor is further configured to modify the grammar information in the dynamic recognition grammar in response to one or more changes to the one or more topological domains associated with the current location associated with the navigation device.

19. The system of claim 12, wherein the dynamic recognition grammar organizes the grammar information according to geographic chunks based on the one or more topological domains associated with the current location associated with the navigation device.

20. The system of claim 19, wherein the conversational language processor is further configured to divide the geographic chunks into multiple tiles, wherein the dynamic recognition grammar further organizes the grammar information according to the multiple tiles associated with the divided geographic chunks.

21. The system of claim 20, wherein the conversational language processor is further configured to divide one or more of the multiple tiles into multiple subtiles, wherein the dynamic recognition grammar further organizes the grammar information according to the multiple subtiles associated with the divided one or more of the multiple tiles.

22. The system of claim 19, wherein the geographic chunks include physical proximities that relate to a distance from the current location associated with the navigation device, temporal proximities that relate to a travel time from the current location associated with the navigation device, directional proximities that relate to a directional travel vector from the current location associated with the navigation device, or civil organizational proximities that relate to continents, countries, regions, states, cities, localities, neighborhoods, or communities mapped to the current location associated with the navigation device.

23. A navigation device for dynamically generating a recognition grammar in an integrated voice navigation services environment, wherein the navigation device comprises one or more processors configured to:

receive a natural language utterance that relates to a navigation context;

identify a current location associated with the navigation device;

generate a dynamic recognition grammar that organizes grammar information based on one or more topological domains associated with the current location associated with the navigation device; and

determine one or more affinities between a user that spoke the natural language utterance and one or more peers associated with the user, wherein the dynamic recognition grammar further organizes the grammar information according to the one or more determined affinities

generate one or more interpretations associated with the natural language utterance using the dynamic recognition grammar.

24. The navigation device of claim 23, wherein the one or more processors are further configured to:

recognize one or more command words in the natural language utterance that define a command in the navigation context;

recognize one or more location words in the natural language utterance that define a state associated with the command in the navigation context, wherein the dynamic recognition grammar further organizes the grammar information according to the state associated with the command in the navigation context; and

recognize one or more additional location words in the natural language utterance that define a city within the state associated with the command in the navigation context, wherein the dynamic recognition grammar further organizes the grammar information according to multiple street addresses in the city within the state associated with the command in the navigation context.

25. The navigation device of claim 24, wherein to generate the one or more interpretations associated with the natural language utterance using the dynamic recognition grammar, the one or more processors are further configured to:

recognize one or more additional command words in the natural language utterance using the grammar information organized according to the multiple street addresses in the city within the state associated with the command in the navigation context, wherein the one or more additional command words define one or more of the multiple street addresses; and

combine the one or more command words and the one or more additional command words recognized in the natural language utterance to generate the one or more interpretations associated with the natural language utterance.

26. The navigation device of claim 23, wherein the one or more topological domains include a state mapped to the current location associated with the navigation device and the one or more processors are further configured to:

recognize one or more command words in the natural language utterance that, define a command in the navigation context; and

recognize one or more location words in the natural language utterance that define a city within the state mapped to the current location associated with the navigation device, wherein the dynamic recognition grammar further organizes the grammar information according to multiple street addresses in the city within the state mapped to the current location associated with the navigation device.

27. The navigation device of claim 26, wherein to generate the one or more interpretations associated with the natural language utterance using the dynamic recognition grammar, the one or more processors are further configured to:

recognize one or more additional command words in the natural language utterance using the grammar information organized according to the multiple street addresses in the city within the state mapped to the current location associated with the navigation device, wherein the one or more additional command words define one or more of the multiple street addresses; and

combine the one or more command words and the one or more additional command words recognized in the natural language utterance to generate the one or more interpretations associated with the natural language utterance.

28. The navigation device of claim 23, wherein the dynamic recognition grammar has a size constrained according to memory or resources available to the navigation device via the integrated voice navigation services environment.

29. The navigation device of claim 23, wherein the conversational language processor is further configured to modify the grammar information in the dynamic recognition grammar in response to one or more changes to the one or more topological domains associated with the current location associated with the navigation device.

30. The navigation device of claim 23, wherein the dynamic recognition grammar organizes the grammar information according to geographic chunks based on the one or more topological domains associated with the current location associated with the navigation device.

31. The navigation device of claim 30, wherein the conversational language processor is further configured to:

divide the geographic chunks into multiple tiles, wherein the dynamic recognition grammar further organizes the grammar information according to the multiple tiles associated with the divided geographic chunks; and

divide one or more of the multiple tiles into multiple subtiles, wherein the dynamic recognition grammar further organizes the grammar information according to the multiple subtiles associated with the divided one or more of the multiple tiles.

32. The navigation device of claim 30, wherein the geographic chunks include physical proximities that relate to a distance from the current location associated with the navigation device, temporal proximities that relate to a travel time from the current location associated with the navigation device, directional proximities that relate to a directional travel vector from the current location associated with the navigation device, or civil organizational proximities that relate to continents, countries, regions, states, cities, localities, neighborhoods, or communities mapped to the current location associated with the navigation device.