Voice recognition system

• US 8,195,461 B2
• Filed: 10/04/2007
• Issued: 06/05/2012
• Est. Priority Date: 12/15/2006
• Status: Active Grant

• Current Assignee: Mitsubishi Electric Corporation
• Original Assignee: Mitsubishi Electric Corporation
• Inventors: Chikuri, Takayoshi
• Primary Examiner(s): ALBERTALLI, BRIAN LOUIS
• Application Number:
  US12/447,352
• Publication Number:
  US 20100076751A1
• Time in Patent Office: 1,706 Days
• Field of Search: None
• US Class Current: 704/252
• 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

A voice recognition system used for onboard equipment having a genre database (DB) that stores search target vocabularies in accordance with respective genres. It has a mike 1 for outputting speech sounds as spoken data; a first voice recognition dictionary 2a for recognizing words of search target genres in the genre DB; a second voice recognition dictionary 2b for recognizing words outside the search target genres; a voice recognition unit 3 for recognizing the speech sounds by collating the spoken data delivered from the mike with the vocabularies contained in the first and second voice recognition dictionaries; an interactive control unit 4 for outputting, when a word delivered from the voice recognition unit as a recognition result is a word obtained using the second voice recognition dictionary, a message so stating as presentation information; and a presentation unit 5 for presenting the presentation information to an outside.

Claims

1. A voice recognition system used for onboard equipment having a genre database that stores search target vocabularies in accordance with respective genres, the voice recognition system comprising:

a microphone for converting speech sounds to an electric signal, and for outputting the electric signal as spoken data;

a first voice recognition dictionary that stores words representing search target genres in the genre database;

a second voice recognition dictionary that stores words not representing search target genres in the genre database;

a voice recognition unit for recognizing the speech sounds by collating the spoken data delivered from the microphone with the words contained in the first voice recognition dictionary and in the second voice recognition dictionary;

a controller programmed to;

when a word delivered from the voice recognition unit as the recognition result indicates one of the genres stored in the first voice recognition dictionary, decide that a search is to be conducted of the target vocabulary corresponding to the recognized genre in the genre database, and

when a word delivered from the voice recognition unit as the recognition result is a word obtained by collating the spoken data with the words contained in the second voice recognition dictionary, generate a message stating that the word is not a word of a search target genre as presentation information; and

a presentation unit for presenting the presentation information delivered from the controller to a user.

2. The voice recognition system according to claim 1, further comprising:

a name database that stores search target vocabularies under respective names; and

a retrieval unit for searching the name database using the word delivered from the controller as a retrieval key, and for sending words which are obtained by the search and include the retrieval key back to the controller, wherein

the controller is further programmed to;

when the word delivered from the voice recognition unit as a recognition result is a word obtained by collating the spoken data with the words contained in the second voice recognition dictionary, initiate the search of the name database by delivering the word to the retrieval unit, and generate as the presentation information the words which includes the retrieval key and are sent back from the retrieval unit in response to the instruction.

3. The voice recognition system according to claim 1, further comprising:

a communication unit for executing a network search using the word delivered from the controller as a keyword, and for sending words including the keyword obtained by the network search back to the interactive control unit, wherein

the controller is further programmed to;

initiate, when the word delivered from the voice recognition unit as a recognition result is a word obtained by collating the spoken data with the words contained in the second voice recognition dictionary, the network search by delivering the word to the communication unit, and outputs

generate as at least part of the presentation information the words which include the keyword and are sent back from the retrieval unit in response to the network search.

4. The voice recognition system according to claim 1, further comprising a dictionary update unit for preparing a new second voice recognition dictionary by replacing the existing second voice recognition dictionary by an update voice recognition dictionary obtained from outside the system.

5. The voice recognition system according to claim 4, further comprising:

a recognition correcting component for issuing an instruction to correct the recognition result delivered from the voice recognition unit; and

a communication unit for transmitting the spoken data delivered from the microphone via the controller to a voice recognition server via a network in response to the instruction from the recognition correcting component, for receiving from the voice recognition server via the network the update voice recognition dictionary generated in a manner as to include a vocabulary used when the voice recognition server is able to recognize the spoken data, and for sending the update voice recognition dictionary back to the controller, wherein

the dictionary update unit acquires the update voice recognition dictionary from the controller, and prepares the new second voice recognition dictionary by replacing the existing second voice recognition dictionary with the update voice recognition dictionary acquired.

6. A voice recognition system for onboard equipment having a genre database that stores search target vocabularies in accordance with respective genres, the voice recognition system comprising:

a microphone for converting speech sounds to an electric signal, and for outputting the electric signal as spoken data;

a first voice recognition dictionary that stores words representing search target genres in the genre database;

a second voice recognition dictionary that stores words not representing search target genres in the genre database;

a voice recognition unit for recognizing the speech sounds by collating the spoken data delivered from the microphone with the words contained in the first voice recognition dictionary and in the second voice recognition dictionary;

an interactive control unit for outputting, when a word delivered from the voice recognition unit as a recognition result is a word obtained by collating the spoken data with the words contained in the second voice recognition dictionary, a message stating that the word is not a word of a search target genre as presentation information;

a presentation unit for presenting the presentation information delivered from the interactive control unit to a user;

a name database that stores search target vocabularies under respective names; and

a retrieval unit for searching the name database using the word delivered from the interactive control unit as a retrieval key, and for sending words which are obtained by the search and include the retrieval key back to the interactive control unit, wherein

the interactive control unit initiates, when the word delivered from the voice recognition unit as a recognition result is a word obtained by collating the spoken data with the words contained in the second voice recognition dictionary, the search by delivering the word to the retrieval unit, and outputs as the presentation information the word which includes the retrieval key and is sent back from the retrieval unit in response to the instruction.

7. A voice recognition method for use with onboard equipment having a genre database that stores search target vocabularies in accordance with respective genres, the method comprising:

using a microphone to convert speech sounds to an electric signal, and output the electric signal as spoken data;

storing in a first voice recognition dictionary words representing search target genres in the genre database;

storing in a second voice recognition dictionary words not representing search target genres in the genre database;

recognizing the speech sounds by collating the spoken data delivered from the microphone with the words contained in the first voice recognition dictionary and in the second voice recognition dictionary;

when the recognition result is a word indicating one of the genres stored in the first voice recognition dictionary, decide that a search is to be conducted of the target vocabulary corresponding to the recognized genre in the genre database, and

when the recognition result is a word obtained by collating the spoken data with the words contained in the second voice recognition dictionary, generate a message stating that the word is not a word of a search target genre as presentation information, and present the presentation information to a user.

8. The method according to claim 7, further comprising:

storing in a name database target vocabularies under respective names; and

searching the name database using the recognized word as a retrieval key to retrieve words that include the retrieval key, wherein

the name database search is initiated when the recognition result is a word obtained by collating the spoken data with the words contained in the second voice recognition dictionary, and the presentation information includes the retrieved words including the retrieval key.

9. The method according to claim 7, further comprising:

executing a network search using the recognition result as a keyword to retrieve words including the keyword, wherein

the network search is initiated when the recognition result is a word obtained by collating the spoken data with the words contained in the second voice recognition dictionary, and

the presentation information includes the retrieved words including the keyword.

10. The method according to claim 7, further comprising:

preparing a new second voice recognition dictionary by replacing the existing second voice recognition dictionary by an update voice recognition dictionary obtained externally.

11. The method according to claim 10, further comprising:

utilizing a recognition correcting component to issue an instruction to correct the recognition result;

transmitting the spoken data delivered from the microphone to a voice recognition server via a network in response to the instruction; and

receiving from the voice recognition server via the network the update voice recognition dictionary, which includes a vocabulary to be used when the voice recognition server is able to recognize the spoken data, wherein

the new second voice recognition dictionary is prepared by replacing the existing second voice recognition dictionary with the update voice recognition dictionary.