Information Retrieving Apparatus, Information Retrieving Method and Navigation System

• US 20120173574A1
• Filed: 09/09/2010
• Published: 07/05/2012
• Est. Priority Date: 09/09/2009
• Status: Active Grant

• Current Assignee: Clarion Co., Ltd. (Faurecia SA)
• Original Assignee: Clarion Co., Ltd. (Faurecia SA)
• Inventors: Homma, Takeshi, Kokubo, Hiroaki, Mizuno, Yoshiyuki
• Granted Patent: US 8,949,133 B2
• Time in Patent Office: Days
• US Class Current: 707/772
• CPC Class Codes: G06F 16/68 Retrieval characterised by using metadata, e.g. metadata not derived from the content or metadata generated manually
• CPC Class Codes: G06F 40/247 Thesauruses; Synonyms
• CPC Class Codes: G10L 15/065 Adaptation
• CPC Class Codes: G10L 15/187 Phonemic context, e.g. pronunciation rules, phonotactical constraints or phoneme n-grams
• CPC Class Codes: G10L 15/19 Grammatical context, e.g. disambiguation of the recognition hypotheses based on word sequence rules
• CPC Class Codes: G10L 15/26 Speech to text systems G10L15/08 takes precedence

Abstract

An information retrieving apparatus is provided which has excellent usability using voice inputs. Search queries input by a user via a touch panel or the like are used as voice recognition vocabulary. Further, the search queries are edited by use of morpheme information and/or other information in a database, thereby providing such voice recognition vocabulary that can be easily uttered by the user.

Claims

1. An information retrieving apparatus comprising:

a name database that registers at least one record as a unit, each of the record contains at least one attribute to be retrieved and a value associated with the attribute, wherein the value is content about each of the attribute;

an operation input unit that receives an operation input of a user;

a voice input unit that receives a voice input of the user;

a voice recognizing unit that recognizes voice obtained from the voice input unit as a character string;

a sound model storing unit that stores sound data referred by the voice recognizing unit;

a language model storing unit that stores, as a language model, a vocabulary recognized by the voice recognizing unit and a connection rule of the corresponding vocabulary;

a retrieving unit that retrieves the value of the attribute included in the name database using an input character string input from the operation input unit or the voice recognizing unit, and creates, as a candidate, the record in which the input character string is included in the value;

an output unit that outputs, as the search result, the candidate of the record created by the retrieving unit;

a selecting unit that selects the output candidate of the record; and

a recognized vocabulary producing unit that receives a selection result of the record by the selecting unit, and produces a new additionally recognized vocabulary that is a voice recognized vocabulary to be added to the language model,

wherein the recognized vocabulary producing unit records, in the name database or the language model, a corresponding relationship between the additionally recognized vocabulary corresponding to the input character string and the selected record.

2. The information retrieving apparatus according to claim 1,

wherein the recognized vocabulary producing unit registers the additionally recognized vocabulary as the single attribute in the record of the name database selected by the selecting means, and adds the additionally recognized vocabulary to the language model storing unit.

3. The information retrieving apparatus according to claim 1,

wherein the additionally recognized vocabulary produced by the recognized vocabulary producing unit is the character string input from the operation input unit.

4. The information retrieving apparatus according to claim 1,

wherein each vocabulary stored in the language model storing unit is stored together with an identifier that indicates the corresponding record of the name database,

wherein the voice recognizing unit outputs the identifier corresponding to the recognized vocabulary together with the recognized vocabulary,

wherein the retrieving unit creates, as a candidate, a record having the same identifier as the name database by referring to the identifier assigned to the vocabulary output by the voice recognizing unit, and

wherein the recognized vocabulary producing unit produces the additionally recognized vocabulary based on the output of the retrieving unit, and adds the additionally recognized vocabulary to the language model storing unit in a form of combining the identifier corresponding to the selected record of the name database with the produced additionally recognized vocabulary.

5. The information retrieving apparatus according to claim 1,

wherein, when the additionally recognized vocabulary is output from the voice recognizing unit, the retrieving unit includes, in the search result, the record of the name database including the additionally recognized vocabulary.

6. The information retrieving apparatus according to claim 1,

wherein, when a plurality of additionally recognized vocabularies is output from the voice recognizing unit, the retrieving unit creates a candidate set of the record by retrieving an additionally recognized vocabulary attribute of the name database based on each of the additionally recognized vocabularies output from the voice recognizing unit, and outputs, as the search result, only a record present in all of the candidate sets of the record that are obtained through search of the respective additionally recognized vocabularies, or outputs, as the search result, collection of records present in at least one candidate set among the candidate sets that are obtained through search of the respective additionally recognized vocabularies.

7. The information retrieving apparatus according to claim 1,

wherein the recognized vocabulary producing unit uses, as the additionally recognized vocabulary, a vocabulary obtained by editing the character string input from the operation input unit using a predetermined method.

8. The information retrieving apparatus according to claim 7,

wherein the recognized vocabulary producing unit uses, as the additionally recognized vocabulary, a vocabulary obtained by editing the character string input from the operation input unit based on a unit of a morpheme of separating the character string that is received in a value of each attribute registered to the name database.

9. The information retrieving apparatus according to claim 1,

wherein the recognized vocabulary producing unit produces the single or the plurality of additionally recognized vocabularies by combining morphemes of separating the character string of a value received in the attribute of the record selected by the selecting means, based on comparison between the character string input from the operation input unit and the selected record.

10. The information retrieving apparatus according to claim 9,

wherein the recognized vocabulary producing unit produces the single or the plurality of additionally recognized vocabularies by combining the morphemes of separating the character string of the value received in the attribute of the record selected by the selecting means, and determines the additionally recognized vocabulary to be adopted based on a type of the morpheme.

13. The information retrieving apparatus according to claim 1, further comprising:

a vocabulary type determining unit that determines a type of the additionally recognized vocabulary produced by the recognized vocabulary producing unit,

wherein the recognized vocabulary producing unit additionally has a function of changing a method of adding the corresponding additionally recognized vocabulary to the name database and the language model storing unit based on the type of the additionally recognized vocabulary determined by the vocabulary type determining unit,

wherein the vocabulary type determining unit compares a value of each attribute in a user selected record that is the record selected by the selecting unit with an input character string that is the character string input from the operation input unit, and outputs, as the determination result, the attribute in the user selected record including the input character string, and

wherein the retrieving unit reads the value of the attribute in the user selected record, retrieves a matching record that is a record matching the value, from the attribute in the name database, and adds the produced additionally recognized vocabulary to an additionally recognized vocabulary

14. The information retrieving apparatus according to claim 1,

wherein the recognized vocabulary producing unit calculates a similarity between the produced additionally recognized vocabulary and the character string stored as the value in the language model storing unit or the attribute of the name database, and enumerates similar vocabularies that are vocabularies similar to the produced additionally recognized vocabulary, and changes a method of adding the produced additionally recognized vocabulary to the name database and the language model when there is a corresponding similar vocabulary.

15. The information retrieving apparatus according to claim 1,

wherein, in addition to the produced additionally recognized vocabulary, the recognized vocabulary producing unit produces a new additionally recognized vocabulary by combining any one attribute of the name database and the additionally recognized vocabulary.

11-12. -12. (canceled)

16-20. -20. (canceled)