Recognition dictionary system structure and changeover method of speech recognition system for car navigation

• US 6,112,174 A
• Filed: 11/13/1997
• Issued: 08/29/2000
• Est. Priority Date: 11/13/1996
• Status: Expired due to Term

• Current Assignee: Renesas Electronics Corporation
• Original Assignee: Hitachi, Ltd., Hitachi Microcomputer System Limited
• Inventors: Tanaka, Makoto, Toge, Tetsuji, Ito, Kouji, Wakisaka, Shinji, Ishiwatari, Kazuyoshi
• Primary Examiner(s): Hudspeth, David R.
• Assistant Examiner(s): Chawan, Vijay
• Application Number:
  US08/970,109
• Time in Patent Office: 1,020 Days
• Field of Search: 704/251, 704/254, 704/231, 704/244, 704/252, 704/243, 704/246, 704/226, 704/233, 704/255, 704/270, 704/275, 704/257, 364/436, 364/449
• US Class Current: 704/251
• CPC Class Codes: G01C 21/3608 using speech input, e.g. using speech recognition
• CPC Class Codes: G10L 2015/228 of application context

Abstract

A speech recognition system realizing large-vocabulary speech recognition at a low cost without deteriorating the rate of recognition and a recognition speed performance is provided with a dictionary change-over section for making a change-over between dictionaries to be subjected to speech recognition in accordance with dictionary change-over information, a first memory for storing a plurality of dictionaries, a second memory for storing one dictionary made an object of recognition, and a speech recognition section for performing a speech recognition processing, whereby speech recognition is performed while making a change-over between dictionaries, as required. For example, in a car navigation speech recognition system, the change-over between dictionaries is made for each area in accordance with position information.

Claims

1. A speech recognition system for a vehicle, comprising:

a first storage section for storing a plurality of dictionaries each of which including a dictionary in which any number or designated number of words or sentences to be made an object of speech recognition are prepared and defined corresponding to a specific one of a plurality of areas on a map;

a second storage section for storing a dictionary selected from said plurality of dictionaries;

a dictionary change-over section for making a change-over between said plurality of dictionaries to select a dictionary from said plurality of dictionaries, in response to input of dictionary change-over information based on position information generated from a Global Positioning System for vehicle navigation;

a speech analysis section for performing a speech analysis processing for an input speech;

a speech recognition section for performing a speech recognition processing for the result of speech analysis by said speech analysis section from the dictionary selected by said dictionary change-over section and stored in said second storage section and an acoustic model; and

a display section for providing a visual display of said dictionary change-over information to inform a user of the location of said vehicle relative to a particular area on the map and the number of words used for said speech recognition is limited to a predetermined number in accordance with a scale of the map displayed on said display section.

2. A speech recognition system according to claim 1, wherein said first storage section includes a memory card and/or a read-only-memory (ROM), and said second storage section includes a random-access-memory (RAM).

3. A speech recognition system according to claim 1, wherein said dictionary change-over information contains position information from said Global Position System for use in a vehicle navigation system for vehicle navigation.

4. A speech recognition system according to claim 1, wherein said dictionary is structured by picking up according to priority among words existed in a corresponding area of said plurality of areas in accordance with scale of the area, said dictionary being composed of names of places, roads, structures, gasoline service stations, convenience stores, and family restaurants, which exist in any or designated area of said plurality of areas, and the number of words of each area being limited to a certain number in accordance with said scale of the area.

5. A speech recognition system comprising:

a first storage section for storing a plurality of dictionaries each of which including a dictionary in which any number or designated number of words or sentences to be made an object of speech recognition are prepared and defined corresponding to a specific one of a plurality of areas;

a second storage section for storing a dictionary selected from said plurality of dictionaries,

a dictionary change-over section for making a change-over between said plurality of dictionaries to select a dictionary from said plurality of dictionaries, in response to reception of dictionary change-over information based on position information generated from a Global Positioning System for vehicle navigation;

a speech analysis section for performing a speech analysis processing for an input speech; and

a speech recognition section for performing a speech recognition processing for the result of speech analysis by said speech analysis section from the dictionary selected by said dictionary change-over section and stored in said second storage section and an acoustic model;

wherein, when any number or designated number of words or sentences made an object of speech recognition are prepared and defined as a dictionary, said dictionary is formed by names of places, roads, structures, gasoline service stations, convenience stores, family restaurants and so forth which exist in any or designated area of said plurality of areas, said dictionary being prepared for each area of said plurality of areas, and related with said acoustic model corresponding to speech recognition for an unspecified speaker.

6. A speech recognition system comprising:

a first storage section including a memory card and/or a read-only-memory (ROM) for storing a plurality of dictionaries each of which including a dictionary in which any number or designated number of words or sentences to be made an object of speech recognition are prepared and defined corresponding to a specific one of a plurality of areas;

a second storage section including a random-access-memory (RAM) for storing a dictionary selected from said plurality of dictionaries;

a dictionary change-over section for making a change-over between said plurality of dictionaries to select a dictionary from said plurality of dictionaries, in response to reception of dictionary change-over information based on position information generated from a Global Positioning System for vehicle navigation;

a speech analysis section for performing a speech analysis processing for an input speech; and

a speech recognition section for performing a speech recognition processing for the result of speech analysis by said speech analysis section from the dictionary selected by said dictionary change-over section and stored in said random-access-memory (RAM) of said second storage section and an acoustic model;

wherein, when any number or designated number of words or sentences made an object of speech recognition are prepared and defined as a dictionary, said dictionary is formed by the names of places, roads, structures, gasoline service stations, convenience stores, family restaurants and so forth which exist in any or designated area of said plurality of areas, said dictionary being prepared for each area of said plurality of areas, and related with said acoustic model corresponding to speech recognition for an unspecified speaker.

7. A speech recognition system comprising:

a first storage section for storing a plurality of dictionaries each of which including a dictionary in which any number or designated number of words or sentences to be made an object of speech recognition are prepared and defined corresponding to a specific one of a plurality of areas,

a second storage section for storing a dictionary selected from said plurality of dictionaries;

a dictionary change-over section for making a change-over between said plurality of dictionaries to select a dictionary from said plurality of dictionaries, in response to reception of dictionary change-over information based on position information generated from a Global Positioning System for vehicle navigation, said dictionary change-over information containing said position information from said Global Positioning System for use in said vehicle for vehicle navigation;

a speech analysis section for performing a speech analysis processing for an input speech; and

a speech recognition section for performing a speech recognition processing for the result of speech analysis by said speech analysis section from the dictionary selected by said dictionary change-over section and stored in said second storage section and an acoustic model;

wherein, when any number or designated number of words or sentences to be made an object of speech recognition are prepared and defined as a dictionary, said dictionary is formed by the names of places, roads, structures, gasoline service stations, convenience stores, family restaurants and so forth which exist in any or designated area of said plurality of areas, said dictionary being prepared for each area of said plurality of areas, and related with said acoustic model corresponding to speech recognition for an unspecified speaker.

8. A speech recognition system for a vehicle, comprising:

a first storage section for storing a plurality of dictionaries each of which including a dictionary in which any number or designated number of words or sentences to be made an object of speech recognition are prepared and defined corresponding to a specific one of a plurality of areas of a map;

a second storage section for storing a dictionary selected from said plurality of dictionaries;

a dictionary change-over section for making a change-over between said plurality of dictionaries to select a dictionary from said plurality of dictionaries, in response to reception of the result of said speech recognition or dictionary change-over information based on position information generated from a Global Positioning System used in a vehicle navigation system;

a speech analysis section for performing a speech analysis processing for an input speech;

a speech recognition section for performing a speech recognition processing for the result of speech analysis from the dictionary selected by said dictionary change-over section and stored in said second storage section and an acoustic model; and

a display section for providing a visual display of said dictionary change-over information to inform a user of the location of said vehicle relative to a particular area on the map and the number of words used for said speech recognition is limited to a predetermined number in accordance with a scale of the map displayed on said display section.

9. A speech recognition system according to claim 8, wherein, when any number or designated number of words or sentences made an object of speech recognition are prepared and defined as a dictionary, said dictionary is formed by the names of places, roads, structures, gasoline service stations, convenience stores, family restaurants and so forth which exist in any or designated area of said plurality of areas, said dictionary being prepared for each area of said plurality of areas to perform speech recognition, and in case where the result of speech recognition a corresponding word does not exist in a dictionary, said speech recognition is performed with a change made to the next dictionary which is made the object of speech recognition.

10. A speech recognition system according to claim 8, wherein, when any number or designated number of words or sentences to be made an object of speech recognition are prepared and defined as a dictionary, said dictionary is formed by the names of places, roads, structures, gasoline service stations, convenience stores, family restaurants and so forth which exist in any or designated area of said plurality of areas, said dictionary being prepared for each area of said plurality of areas to perform speech recognition, and in case where the result of speech recognition indicates an index of a dictionary, said speech recognition is performed with a change made to a dictionary as an object of speech recognition which is indicated by the index.

11. A speech recognition system according to claim 8, wherein said dictionary is structured by picking up according to priority among words existed in a corresponding area of said plurality of areas in accordance with scale of the area, said dictionary being composed of names of places, roads, structures, gasoline service stations, convenience stores, and family restaurants, which exist in any or designated area of said plurality of areas, and the number of words of each area being limited to a certain number in accordance with said scale of the area.

12. A speech recognition system according to claim 8, wherein said dictionary is composed of names of places, roads, structures, gasoline service stations, convenience stores, and family restaurants which exist in any or designated area of said plurality of areas, and the number of words of each area being limited to a certain number.

13. A speech recognition system for vehicle navigation using a Global Positioning System, comprising:

a first storage unit which stores a plurality of dictionaries, each dictionary indicating a list of directory information of a particular area on a map, including names of roads, places, structures and streets, street numbers, names of intersections, gasoline service stations, convenience stores, and/or family restaurants;

a second storage unit which stores a dictionary selected from said plurality of dictionaries for speech recognition;

a dictionary change-over unit which makes a change-over between different dictionaries stored in said first storage unit for selecting a dictionary from said plurality of dictionaries for speech recognition, in response to input of dictionary change-over information generated from said Global Positioning System for said vehicle navigation;

a speech analysis unit which performs a speech analysis of an input speech from a user requesting said vehicle navigation;

a speech recognition unit which performs a speech recognition after said speech analysis based on said dictionary selected from said plurality of dictionaries by said dictionary change-over unit and stored in said second storage unit and an acoustic model for said vehicle navigation; and

a display unit for providing a visual display of said dictionary change-over information to inform a user of the location of said vehicle relative to said particular area on the map and the number of words used for said speech recognition is limited to a predetermined number in accordance with a scale of the map displayed on said display unit.

14. A speech recognition system according to claim 13, wherein said first storage unit contains a memory card and/or a read-only-memory (ROM), and said second storage unit contains a random-access-memory (RAM).

15. A speech recognition system according to claim 13, wherein said dictionary change-over information contains location information of said vehicle relative to a particular area on the map.

16. A speech recognition system according to claim 15, further comprising a display unit for providing a visual display of said dictionary change-over information to inform a user of the location of said vehicle relative to said particular on the map.

17. A speech recognition system according to claim 13, wherein said dictionary change-over information is periodically updated by said Global Positioning System for tracking the location of said vehicle along the map.

18. A speech recognition system according to claim 13, wherein said first storage unit further stores an address book dictionary which includes names, address, and phone numbers of registered persons.

19. A speech recognition system according to claim 18, wherein said speech recognition unit accesses said address book dictionary stored in said first storage unit to recall information of a designated name for vehicle navigation.

20. A speech recognition system according to claim 19, wherein said dictionary selected is composed of names of places, roads, structures, gasoline service stations, convenience stores, and family restaurants which exist in said particular area on the map.

21. A speech recognition system, comprising:

a storage unit which stores a plurality of dictionaries, each dictionary containing a collection of words or sentences prepared for speech recognition for providing information of a specific area on a map;

a speech analysis unit which analyzes an input speech from a user;

a speech recognition unit which recognizes said input speech based on speech analysis and an acoustic model for allowing access to information of a specific area on the map contained in a selected dictionary;

a dictionary change-over unit which makes a change-over between different dictionaries for said selected dictionary in response to input of dictionary change-over information; and

display unit for providing a visual display of said dictionary change-over information to inform location of a user relative to said specific area on the map, and the collection of words or sentences used for said speech recognition is limited to a predetermined number in accordance with a scale of the map displayed on the display unit.

22. A speech recognition system according to claim 21, wherein said storage unit contains a read-only-memory (ROM) for storing said plurality of dictionaries, and a random-access-memory (RAM) for storing said selected dictionary.

23. A speech recognition system according to claim 21, wherein said dictionary change-over information contains location information of a vehicle relative to said specific area on the map, when said speech recognition system is incorporated into a vehicle navigation system using a Global Positioning System.

24. A speech recognition system according to claim 23, wherein said dictionary change-over information is periodically updated by said Global Positioning System for tracking the location of said vehicle along the map.

25. A speech recognition system according to claim 21, wherein said selected dictionary is composed of names of places, roads, structures, gasoline service stations, convenience stores, and family restaurants which exist in said specific area on the map.

26. A speech recognition system according to claim 21, wherein said storage unit further stores an address book dictionary which includes names, address, and phone numbers of registered persons, and wherein said speech recognition unit accesses said address book dictionary stored in said storage unit to recall information of a designated name for vehicle navigation.