Method for searching data in at least two databases

• US 7,363,222 B2
• Filed: 06/24/2002
• Issued: 04/22/2008
• Est. Priority Date: 06/28/2001
• Status: Expired due to Fees

• Current Assignee: Nokia Corporation
• Original Assignee: Nokia Corporation
• Inventors: Josenhans, Michael
• Primary Examiner(s): Lerner; Martin
• Application Number:
  US10/482,517
• Publication Number:
  US 20040243562A1
• Time in Patent Office: 2,129 Days
• Field of Search: 704/231, 704/236, 704/238, 704/251, 704/270.1, 704/275, 707/1, 707/2, 707/3, 707/104.1
• US Class Current: 704/231
• CPC Class Codes: G10L 15/26 Speech to text systems G10L15/08 takes precedence
• CPC Class Codes: H04M 1/271 controlled by voice recognition
• CPC Class Codes: H04M 1/275 implemented by means of portable electronic directories
• CPC Class Codes: H04M 2250/02 including a Bluetooth interface
• CPC Class Codes: Y10S 707/99933 Query processing, i.e. searching

Abstract

A method and database system is disclosed for searching data in at least two databases (D_n), particularly for searching telephone directories or the like. To allow simultaneous access to two or more databases by means of speech recognition in order to perform a search therein as in a single database, a search term is input by speech via a voice controlled user interface (28) connected to a database primary control apparatus (26) and comprises speech recognition front end means (8, 9) for processing a sound sequence of a search term input by speech to obtain a comparable speech pattern (X) thereof. By means of speech recognition back end means (6) associated with databases (D₁-D₆), the comparable speech pattern (X) is compared with corresponding speech patterns (A_n,i) of database entries (E_n,i) to determine for each of the at least two databases (D_n) at least that database entry (E_n,j) the speech pattern (A_n,j) which best matches the comparable speech pattern (X) of the search term.

Claims

1. A method for searching data in at least two databases comprising:

inputting a search term by speech,

processing a sound sequence of the search term to obtain a comparable speech pattern thereof,

comparing the comparable speech pattern of the search term with corresponding speech patterns of database entries to determine for each entry of the at least two databases a measure for the matching distance between the compared speech patterns to determine at least that database entry of each database the speech pattern of which matches the comparable speech pattern of the search term best,

selecting that database entry the speech pattern of which matches the comparable speech pattern of the search term best from the entries determined previously by searching that measure that indicates the best matching speech pattern of the database entries, and

outputting an identifier of that database entry that has been selected previously,

wherein the identifier of that database entry is entered as a search term for an alphanumeric search through the at least two databases, wherein

on the basis of the entered search term, the predecessor or a plurality of predecessors and the successor or a plurality of successors to the search term and, if available, the search term itself from each of the databases are stored in a search table memory, and

the search term, if stored in the search table memory, or that term from the successors stored in the search table memory which comes closest after the entered search term, is displayed.

2. A method according to claim 1, wherein for each database a respective table is set up that contains all database entries in the order in which their speech pattern phonetically match the speech pattern of the search term.

3. A method according to claim 1, wherein processing the sound sequence of the search term results in a phoneme sequence used as comparable speech pattern.

4. A method according to claim 1, wherein the measure that indicates the closest matching distance between the compared speech patterns is determined in accordance with the following equation:

m_n(X,A_n,j)<

m(X,A_n,i) for all i≠

J

wherein i =1, 2, . . . k_n and k_n is the number of entries of n^th database and m is the measure.

5. A method according to claim 1, wherein the measures indicating the best matching database entries of each database are compared to select that database entry the speech pattern of which matches the comparable speech pattern of the search term best.

6. A method according to claim 1, wherein

the comparable speech pattern of the search term is compared with the comparable speech patterns of those database entries that have been determined as that of the respective databases that match the search term best to determine for each of these entries a measure for the matching distance between the compared speech patterns, and

at least that database entry that matches the search term best of all is determined by searching that measure that indicates the best matching speech pattern of all the database entries of all databases.

7. A method according to claim 6, wherein the measure that indicates the closest matching distance between the compared speech patterns is determined in accordance with the following equation:

m_h(X,A_p,j)<

m(X,A_n,j) for all n≠

p

wherein n =1, 2, . . . q and q is the number of databases to be searched.

8. A device for searching data in at least two databases comprising:

means for inputting a search term by speech,

means for processing a sound sequence of the search term to obtain a comparable speech pattern thereof,

means for comparing the comparable speech pattern of the search term with corresponding speech patterns of database entries to determine for each entry of the at least two databases a measure for the matching distance between the compared speech patterns to determine at least that database entry of each database the speech pattern of which matches the comparable speech pattern of the search term best,

means for selecting that database entry the speech pattern of which matches the comparable speech pattern of the search term best from the entries determined previously by searching that measure that indicates the best matching speech pattern of the database entries, and

means for outputting an identifier of that database entry that has been selected previously,

wherein the identifier of that database entry is entered as a search term for an alphanumeric search through the at least two databases,

means for storing, on the basis of the entered search term, the predecessor or a plurality of predecessors and the successor or a plurality of successors to the search term and, if available, the search term itself from each of the databases, in a search table memory, and

means for displaying the search term, if stored in the search table memory, or that term from the successors stored in the search table memory which comes closest after the entered search term.

9. A device for searching data in at least two databases comprising:

a handset for inputting a search term by speech,

a speech recognition front end module for processing a sound sequence of the search term to obtain a comparable speech pattern thereof,

a speech recognition back end module for comparing the comparable speech pattern of the search term with corresponding speech patterns of database entries to determine for each entry of the at least two databases a measure for the matching distance between the compared speech patterns to determine at least that database entry of each database the speech pattern of which matches the comparable speech pattern of the search term best,

a speech recognition middle part module for selecting that database entry the speech pattern of which matches the comparable speech pattern of the search term best from the entries determined previously by searching that measure that indicates the best matching speech pattern of the database entries, and

a user interface for outputting an identifier of that database entry that has been selected previously,

wherein the identifier of that database entry is entered as a search term for an alphanumeric search through the at least two databases,

a memory for storing, on the basis of the entered search term, the predecessor or a plurality of predecessors and the successor or a plurality of successors to the search term and, if available, the search term itself from each of the databases, in a search table memory, and

a display for displaying the search term, if stored in the search table memory, or that term from the successors stored in the search table memory which comes closest after the entered search term.