GB2463371A - Retrieving route information using speech recognition and spoken postal codes - Google Patents
Retrieving route information using speech recognition and spoken postal codes Download PDFInfo
- Publication number
- GB2463371A GB2463371A GB0915921A GB0915921A GB2463371A GB 2463371 A GB2463371 A GB 2463371A GB 0915921 A GB0915921 A GB 0915921A GB 0915921 A GB0915921 A GB 0915921A GB 2463371 A GB2463371 A GB 2463371A
- Authority
- GB
- United Kingdom
- Prior art keywords
- character
- portions
- sub
- whole
- sound
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
- 230000005236 sound signal Effects 0.000 claims description 14
- 230000001131 transforming effect Effects 0.000 claims description 4
- 238000000034 method Methods 0.000 description 14
- 238000001514 detection method Methods 0.000 description 4
- 238000005516 engineering process Methods 0.000 description 3
- 238000010586 diagram Methods 0.000 description 2
- 230000000694 effects Effects 0.000 description 1
- 239000004973 liquid crystal related substance Substances 0.000 description 1
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01C—MEASURING DISTANCES, LEVELS OR BEARINGS; SURVEYING; NAVIGATION; GYROSCOPIC INSTRUMENTS; PHOTOGRAMMETRY OR VIDEOGRAMMETRY
- G01C21/00—Navigation; Navigational instruments not provided for in groups G01C1/00 - G01C19/00
- G01C21/26—Navigation; Navigational instruments not provided for in groups G01C1/00 - G01C19/00 specially adapted for navigation in a road network
- G01C21/34—Route searching; Route guidance
- G01C21/36—Input/output arrangements for on-board computers
-
- G—PHYSICS
- G10—MUSICAL INSTRUMENTS; ACOUSTICS
- G10L—SPEECH ANALYSIS TECHNIQUES OR SPEECH SYNTHESIS; SPEECH RECOGNITION; SPEECH OR VOICE PROCESSING TECHNIQUES; SPEECH OR AUDIO CODING OR DECODING
- G10L15/00—Speech recognition
-
- G—PHYSICS
- G10—MUSICAL INSTRUMENTS; ACOUSTICS
- G10L—SPEECH ANALYSIS TECHNIQUES OR SPEECH SYNTHESIS; SPEECH RECOGNITION; SPEECH OR VOICE PROCESSING TECHNIQUES; SPEECH OR AUDIO CODING OR DECODING
- G10L15/00—Speech recognition
- G10L15/22—Procedures used during a speech recognition process, e.g. man-machine dialogue
-
- G—PHYSICS
- G10—MUSICAL INSTRUMENTS; ACOUSTICS
- G10L—SPEECH ANALYSIS TECHNIQUES OR SPEECH SYNTHESIS; SPEECH RECOGNITION; SPEECH OR VOICE PROCESSING TECHNIQUES; SPEECH OR AUDIO CODING OR DECODING
- G10L15/00—Speech recognition
- G10L15/26—Speech to text systems
-
- G—PHYSICS
- G10—MUSICAL INSTRUMENTS; ACOUSTICS
- G10L—SPEECH ANALYSIS TECHNIQUES OR SPEECH SYNTHESIS; SPEECH RECOGNITION; SPEECH OR VOICE PROCESSING TECHNIQUES; SPEECH OR AUDIO CODING OR DECODING
- G10L15/00—Speech recognition
- G10L15/22—Procedures used during a speech recognition process, e.g. man-machine dialogue
- G10L2015/226—Procedures used during a speech recognition process, e.g. man-machine dialogue using non-speech characteristics
- G10L2015/228—Procedures used during a speech recognition process, e.g. man-machine dialogue using non-speech characteristics of application context
Landscapes
- Engineering & Computer Science (AREA)
- Radar, Positioning & Navigation (AREA)
- Remote Sensing (AREA)
- Physics & Mathematics (AREA)
- Human Computer Interaction (AREA)
- Computational Linguistics (AREA)
- Health & Medical Sciences (AREA)
- Audiology, Speech & Language Pathology (AREA)
- Acoustics & Sound (AREA)
- Multimedia (AREA)
- General Physics & Mathematics (AREA)
- Automation & Control Theory (AREA)
- Traffic Control Systems (AREA)
- Navigation (AREA)
- Character Discrimination (AREA)
Abstract
The utterance of the first half phrase of a postal code number of Britain is urged (S210). Speech recognition is carried out (S220). If the result is "CV2," the numeral portion in the second half phrase is designated so as to be able to be connected with "CV2" (S230). A dummy recognition dictionary for the second half phrase is prepared by enumerating all the combinations with respect to the designated numeral portion (S240). The utterance of the second half phrase is urged (S250). An input speech is recognized by collating with the phrases in the dictionary (S260). If "2TB" is uttered, it is recognized as "2TB." The postal code number is recognized based on the recognized first half phrase and the numeral portion of the recognized second half phrase (S270). If the first half phrase is CV2 and the second half phrase is2TB, "CV2-2"is recognized as the postal codenumber.
Description
CODE RECOGNITION APPARATUS AND ROUTE RETRIEVAL APPARATUS
FIELD OF THE INVENTION
The present invention relates to a code recognition apparatus and a route retrieval apparatus.
BACKGROUND OF THE INVENTION
[Patent document 1] JP-H5-019684 A Conventionally, an apparatus is known which specifies a spot on a map in response to an input of a postal code number (for example? see Patent document 1). If such an apparatus is configured as a car navigation apparatus, it is convenient for a user to input a postal code number using speech recognition. The following explains how to recognize a postal code number for designating a spot using such a conventional technology by taking an example in Britain.
The postal code number in Britain is defined in accordance with the following rule.
The rule defines a postal code number as consisting of a first half phrase and a second half phrase, each phrase containing a numerical character and an alphabetical character such as "LE67-2DQ." Furthermore, in the first half phrase, the character number of numeral characters or alphabetical characters changes depending on areas in Britain.
In contrast, in the second half phrase, the character number of numeral characters or alphabetical characters is constantly fixed; namely, the second half phrase consists of a single numeral character and two alphabetical characters.
Incidentally, the map data actually include data of postal code numbers as consisting of (i) the first half phrase plus (ii) a numeral character portion of the second half phrase. For example, the postal code number "LE67-2DQ" is included in the map data as "LE67-2." Therefore, a route retrieval apparatus is required to recognize [first half phrase + the numeral character portion of the second half phrase] to designate a destination. Furthermore, in the present state, map information associating postal code numbers and spots with each other is not generally obtained in Britain.
The following explains a conventional speech recognition processfor recognizing such [the first half phrase + a numeral character portion of the second half phrase] using FIG. 3. FIG 3 shows a conventional speech recognition process executed by a computer in a route retrieval apparatus. First, the user is prompted to utter the first half phrase of a postal code number via a speaker (SIlO).
Next, speech recognition of the phrase is carried out with respect to the first half phrase via a microphone (S120). That is, based on a sound signal showing the user's speech utterance inputted via the microphone, a character string corresponding to the first half phrase of the postal code number is recognized using a recognition dictionary.
The following exemplifies a case that the result is "CV2." A numeral character portion of the second half phrase is designated which succeeds "CV2" with reference to the information on postal code number stored in the map data (S130). For example, the result of the designation is exemplified as "2,4, 7, and 9." Based on such a result of the designation, a recognition dictionary is prepared (S140). That is, the recognition dictionary is prepared in association with sound data representing the sound characteristic of each numeral' character of "2, 4, 7, and 9.
Next, the user is prompted via the speaker to utter the second half phrase of the postal code number (5150). The sound signal represe.nting the user's speech utterance inputted via the microphone and the sound data in the recognition dictionary are collated with each other. The numeral character portion of the second half phrase is thereby recognized (S 160). For example, if the user utters "2TB," it is recognized as "2." Finally, the postal code number is recognized based on the result of the speech recognition (8170). In the foregoing exemplified case, the recognition is "CV2-2.' The reason for dividing the postal code number into the halves to thereby recognize is to improve the recognition rate. That is, with respect to the second half phrase, a recognition dictionary containing a small number of candidates can be prepared based on the recognition result of the first half phrase. The recognition rate thus becomes improved.
The foregoing technology has a technical problem in that the apparatus needs to recognize "2" although the user utters "2TB." In such a case, the excessive infomiation of "TB" sometimes causes the recognition of important M2h1 to fail. The sound signal of 2TB" is collated with the sound data for speech recognition in thedictionary corresponding to only "2" Impossibility may eventually arise in determining a matching degree.
SUMMARY OF THE INVENTION
In consideration of the above mentioned technical problem, it is an object to provide a code recognition apparatus to more securely perform speech recognition of a speech containing excessive information, and a route retrieval apparatus using the foregoing code recognition apparatus.
According to an example of the present invention, a code recognition apparatus is provided as follows. The apparatus is for recognizing a code among codes of a specific kind, each of the codes satisfying a predetermined rule of the specific kind and being a character string consisting of a predetermined first portion and a predetermined second portion. The second portion is also a character string. The apparatus comprises: a sound collecting means for collecting a sound in a surrounding area and transforming the collected sound to thereby generate a sound signal; a first recognizing means for recognizing a first portion of the codes of the specific kind based on the sound signal, generated by the sound collecting means; an enumerating means for enumerating a whole of character strings of second portions, wherein the predetermined rule is satisfied when each of the whole of character strings is united with the first portion recognized by the first recognizing means; a dictionary preparing means for preparing a speech recognition dictionary in which the whole of the character strings of the second portions enumerated by the enumerating means and sound data elements are respectively associated with each other a second recognizing meansfor recognizing, as a character string of a second portion, a character stung corresponding to a sound data element most matching with a sound signal generated by the sound collecting means, from among the sound data elements included n the speech recognition dictionary after the dictionary preparing means prepares the speech recognition dictionary; and an extracting means for extracting a predetermined sub portion, which is included in the second portion, from the character string recognized by the second recognizing means.
As another example of the present invention, a route retrieval apparatus is provided as comprising the above-recited code recognition apparatus where the specific kind is a postal code number. The route retrieval apparatus further comprises a retrieval means for retrieving a route between spots. Herein, the database includes a spot data element with respect to each of the character strings consisting of (i) the preceding portions and (ii) sub portions of the second portions succeeding the preceding portions; the acquiring means acquires a spot data element from the database, based on a character string of the preceding portion recognized by the first recognizing means and a sub portion of the succeeding portion extracted by the extracting means, the sub portion directly succeeding the preceding portion; and the retrieving means designates, as a start point or an end point of a route to be retrieved, a spot indicated by the spot data element acquired by the acquiring means.
BRIEF DESCRIPTION OF THE DRAWINGS
The above and other objects, features, and advantages of the present invention will become more apparent from the following detailed description made with reference to the accompanying drawings. In the drawings: FIG. I is a diagram illustrating a schematic configuration of a route retrieval apparatus according to an embodiment of the present invention; FIG. 2 is a flowchart illustrating a route retrieval process; and FIG. 3 is a flowchart illustrating a conventional speech recognition process.
DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS
FIG. I is a schematic block diagram of a route retrieval apparatus in a vehicle according to an embodiment of the present invention, The route retrieval apparatus 10 includes the following: a position detection section 11 for detecting a present position of the vehicle; an operation switch group 12 functioning as a user interface; a sound input section 13 for receiving an input of a sound or speech; a map data input section 17 for reading map data etc. from a storage medium; a display section 14 for displaying a map, a route guidance window, etc.; a sound output section 15 for outputting various kinds of guidance sounds; and a controller 20. The controller 20 executes various processes based on inputs from the position detection section 11, the operation switch group 12, the sound input section 13, and the map data input section 17. The controller 20 controls each section of the route retrieval apparatus 10.
The position detection section Ii includes the following: a GPS receiver ha which receives via a GPS antenna (not shown) electric waves from satellites for GPS (Global Positioning System) and obtains orbital information on satellites and information on present date and time; a gyroscope lib for detecting rotational movement exerted to the vehicle; a distance sensor lic which measures a travel distance of the vehicle; and a geomagnetic sensor lid which measures a heading direction of the vehicle from earth mag netisrn.
The operation switch group 12 includes a mechanical key switch arranged in the circumference of the display section 14 and a touch sensitive panel laminated on and integrated into a surface of the screen on the display section 14. In addition, the Sound input section 13 includes a microphone for collecting surrounding sounds and transforming the collected sounds into sound signals, so as to obtain speech information which the user utters.
The map data input section 17 is used for inputting to the controller 20 map data stored in a storage medium (unshown). The storage medium stores as map data postal code number data, node and link data indicating positions and connecting relations of roads, topographical data, and facility data. In addition, data for map matching for improving the position detection accuracy, and sound data for route guidance, etc. are stored. A CD-ROM, a DVD, and a hard disk can be exemplified as a storage medium.
Furthermore, the route retrieval apparatus 10 according to the present embodiment is assumed to be used in Britain. The map data thus indicates geo'graphy of Britain.
The display section 14 is a color display device configured of a liquid crystal display monitor etc. The display section 14 is caused by the controller 20 to display a map around a present position of the vehicle, or a route to a destination designated by a user, etc. In addition, the sound output section 15 is configured of a speaker etc. and is caused by the controller 20 to output guidance sounds concerning route guidance to a destination.
In addition, the controller 20 is configured of a CPU 20a, a RAM 20b, a ROM 20c, a NVRAM 20d, etc. The CPU 20a executes various programs stored in the ROM 20c.
While carrying out overall control of each section of the apparatus 10, the CPU 20a executes a map display function, a route retrieval function, a route guidance function, or a speech recognition function.
Next, the route retrieval process is explained using FK3. 2. The route retrieval process is executed mainly by the controller 20. In addition, the route retrieval process is started by a trigger signal inputted via the sound input section 13. In particular, the route retrieval process is executed when the user inputs via the sound input section 13 an intension to input by speech a postal code number as information to designate a position such as a destination necessary for the route retrieval.
It is noted that S210 and SIlO, S220 and S120, and S230 and S130 are identical to each other, respectively; thus, the explanation is omitted. The explanation starts with S240. A dummy recognition dictionary is prepared by enumerating all the character strings, which can be combined with the numeral character portion in the second half phrase designated at S230 (S240).
In detail, in accordance with the rule of the postal code number of Britain, the whole of possible combinations of two alphabetical characters, for instance, AA, AB, AC, ..., ZX, ZY, and ZZ, are added to the designated numeral character portion. The character strings of the patterns of 676 (square of 26) are prepared with respect to each designated numeral character portion of the second half phrase. The Sound data element corresponding to each character string is then prepared. A dummy recognition dictionary is p!epared which associates each character string with the sound data element.
It is noted that the sound data element represents a feature of the speech of the corresponding phrase. The sound data element may represent sound waves of the corresponding phrase or represent a feature quantity of the speech.
The following summarizes the processing at S230 and S240. The storage medium set to the map data input section 17 stores the postal code number data. In the postal code number data, character strings and position information items are respectivelyassociated with each other. Every character string as a combination of characters contains a part of a postal code number and consists of (i) a first half portion or phrase (i.e., a preceding portion or phrase) of the postal code number and (ii) a numeral character portion or phrase of a second half portion (i.e., a succeeding portion or phrase) of the postal code number. The position information item indicates a position (latitude and longitude) of a spot specified by the associated character string. At S230, from a group of "a part of a postal code number" registered in the storage medium, all the character strings that contain as the first half phrase the character string recognized at S220 are retrieved. It is noted that each of the retrieved character strings include the first half phrase and a numeral character portion of the second half phrase.
At S240, first, the whole of combinations of the two alphabetical characters are enumerated (i.e., listed). Second, all the combinations of (i) the numeral character portion, which is included in each character string obtained by the retrieval at S230, and (ii) the above enumerated two alphabetical characters are enumerated. A dummy recognition dictionary is then prepared In which each character string of each of above-mentioned all the enumerated combinations is associated with a sound data element.
Next, the user is prompted to utter the second half portion of the postal code number via the sound output section 15 (S250). The user's speech inputted via the sound input section 13 is recognized by collating with the sound data elements in the prepared dummy recognition dictionary (S260). For instance, if U2TB is spoken or pronounced by the user, it is recognized as "2TB." That is, the speech recognition process searches the dummy recognition dictionary for the sound data element which matches most with the feature of the sound signal representing the users speech inputted via the sound input section 13. The process obtains information of the characterstring most matched with from the dummy recognition dictionary to thereby recognize the character string of the three characters uttered by the user.
The leading numeral character is extracted from the three characters of the second half portion of the postal code number. Thus, as a format stored in the map data, the first half phrase and the recognized numeral character portion of the second half phrase (S270). Herein, the numeral character portion of the second half phrase or portion may be referred to as a sub portion of the second half phrase or portion. For instance, if the first half phrase is CV2 and the second half phrase is 2TB, "CV2-2" is recognized as the postal code number.
The process specifies a spot corresponding to the part of the postal code number recognized at S270 (S280). That is, at S280, the process obtains or acquires a position information item, which is associated with the character string recognized as a part of the postal code number at S270, from the above storage medium. The spot indicated by the obtained position information item is specified as a spot corresponding to the part of the postal code number recognized at S270..
While designating the spot specified at S280 as a destination, the process designates the present position as. a departure point to thereby retrieve an optimal route from the departure point to the destination. The optimal route retrieved is presented to the user via the display section 14 (S290).
The foregoing describes the embodiment of the present invention. The present embodiment can provide an advantage as follows. A dummy recognition dictionary is prepared to record sound data elements of all the combinations of character strings, which the user is expected to utter or speak. Based on the dummy recognition dictionaiy, the userts speech is recognized. The speech recognition rate is thus improved.
The following summarizes a featured configuration of the present embodiment.
That is, the recognition dictionary is not prepared only with respect to the information needed. The recognition dictionary is prepared with respect to all the possible speeches, which the user is expected to utter, although the recognition dictionary contains information useless in regard to the recognition of the postal code number in the map data. The useless information is eventually deleted. In other words, the present embodiment includes a step of deleting the useless information eventually. It is noted that preparing of all the combinations is necessary since the rule of the character strings of the second half portion is known but it is difficult to obtain, as map data, such character strings according to the rule. Thus, the present embodiment can provide an effect peculiar to the present invention by executing the step which superficially seems to be excessive.
Furthermore, the present invention is not limited to the above embodiment. For example, in the route retrieval, both a departure point and a destination may be obtained from a user using the speech recognition of postal code numbers.
(Functions) Each step executed by the controller 20 may function as means or section as follows. S220 may function as a first recognizing means or section. S230 and S240 may function as a listing or enumerating means or section. S240 may function as a dictionary preparing means or section. S260 may function as a second recognizing means or section. S270 may function as an extracting means or section. S280 may function.as an acquiring means or section. S290 may function as a retrieving means or section.
Each or any combination of processes, steps, or means explained in the above can be achieved as a software section or unit (e.g., subroutine) and/or a hardware section or unit (e.g., circuit or integrated circuit), including or not including a function of a related device; furthermore, the hardware section or unit can be constructed inside of a microcomputer.
Furthermore, the software section or unit or any combinations of multiple software sections or units can be included in a software program, which can be contained in a computer-readable storage media or can be downloaded and installed in a computer via a communications network.
Aspects of the disclosure described herein are set out in the following clauses.
The following listing of claims will replace all prior versions, and listings, of claims in the present application.
As an aspect of the disclosure, a code recognition apparatus is provided as follows. The apparatus is for recognizing a code among codes of a specific kind, each of the codes satisfying a predetermined rule of the specific kind and being a character string consisting of a predetermined first portion and a predetermined second portion.
The second portion is also a character string. The apparatus comprises: a sound collecting means for collecting a sound in a surrounding area and transforming the collected sound to thereby generate a sound signal; a first recognizing means for recognizing a first portion of the codes of the specific kind based on the sound signal generated by the Sound collecting means; an enumerating means for enumerating a whole of character strings of second portions, wherein the predetermined rule is satisfied when each of the whole of character strings is united with the first portion recognized by the first recognizing means; a dictionary preparing means for prepanng a speech recognition dictionary in which the whole of the character strings of the second portions enumerated by the enumerating means and sound data elements are respectively associated with each other; a second recognizing means for recognizing, as a character string.of a second portion, a character string corresponding to a sound data element most matching with a sound signal generated by the Sound collecting means, from among the sound data elements included in the speech recognition dictionary after the dictionary preparing means prepares the speech recognition dictionary; and an extracting means for extracting a predetermined sub portion, which is included in the second portion, from the character string recognized by the second recognizing means.
According to the above code recognition apparatus, the recognition of the speech containing excessive information can be executed more certainly. This is because the inputted speech is once recognized as it is, instead of recognizing a part of the inputted speech like a conventional technology. The required portion is then extracted by deleting the excessive information.
For instance, "LE67-2DQ" is uttered. In such a case, the whole of "LE672DQ" is recognized as it is. The form of "LE67-2" is then prepared from the recognized whole of "LE67-2DQ by extracting a part of the second half portion. Thus, the uttered speech can be more securely speech recognized and only the required portion can be obtained among the information which has been recognized.
As an optional aspect, the code recognition apparatus may further comprise a storing means and an acquiring means. The storing means is for storing a database containing a data element corresponding to each of the codes of the specific kind. The data element is associated with each combination consisting of (i) a character string of a first portion to be recognized by the first recognizing means and (ii) a character of a sub portion in a second portion to be extracted by the extracting means. The acquiring means is for acquiring, from the database, a data element associated with each combination consisting of (i) the character string of the first portion recognized by the first recognizing means and (ii) the character of the sub portion in the second portion extracted by the extracting means.
Thus, in cases that a sound input contains both the information necessary for retrieval of the database and the excessive infqrmation unnecessary for the retrieval, the above configuration can provide a more secured retrieval of the database.
As a further optional aspect, in the above code recognition apparatus, each of the codes of the specific kind may be a postal code number which consists of a combination of a character string of a preceding portion and a character string of a succeeding portion. The first portion may be the preceding portion of the postal code number. The second portion may be the succeeding portion of the postal code number The sub portion in the second portion may be a part of the succeeding portion of the postal code number, the sub portion in the second portion succeeding the preceding portion of the postal code number.
The above configuration can enable a more certain acquisition of a data element associated with a character string consisting of a preceding half portion and a sub portion of a succeeding half portion, for instance, with respect of a postal code number.
As a yet further optional aspect, in the code recognition apparatus, the first portion may consist of an optional combination of a numeraF character and an alphabetical character. The second portion may consist of a single numeral character and two alphabetical characters succeeding the single numeral character. The sub portion may be a single numeral character included in the second pOrtion. The enumerating means may (i) retrieve a whole of character strings, each of which includes the first portion recognized by the first recognizing means as a first portion of a postal code number, from a group of character strings respectively consisting of first portions and sub portions, the group of the character strings being registered in the database as combinations of the first portions and the sub portions, (ii) designate a whole of numeral characters consisting sub portions of the retrieved whole of character strings; (iii) enumerate, with respect to each of the designated whole Of numeral characters, a whole of combinations of two alphabetical characters succeeding the each of the designated whole of numeral characters; and (iv) form character strings constituting second portions by combining the each of the designated whole of numeral characters and the enumerated whole of combinations of two alphabetical characters.
The above configuration can enable a more certain acquisition of a data element associated with a character string consisting of a preceding half portion. and a sub portion of a succeeding half portion, for instance, with respect of a postal code number in map data of Britain. It is because the rule of the postal code number of Britain agrees with the above configuration.
As another aspect of the disclosure, a route retrieval apparatus is provided as compnsing the above-recited code recognition apparatus where the specific kind is a postal code number. The route retrieval apparatus further comprises a retrieval means for retrieving a route between spots. Herein, the database includes a spot data element with respect to each of the character strings consisting of (I) the preceding portions and (ii) sub portions of the second portions succeeding the preceding portions; the acquiring means acquires a spot data element from the database, based on a character string of the preceding portion recognized by the first recognizing means and a sub portion of the succeeding portion extracted by the extracting means, the sub portion directly succeeding the preceding portion; and the retrieving means designates, as a start point or an end point of a route to be retrieved, a spot indicated by the spot data element acquired by the acquiring means.
In the above configuration, in specifying a spot using a postal code number at the time of a route retrieval, even when a postal code number is inputted by speech as including characters unnecessary for retrieving the spot, the erroneous recognition of the spot can be avoided and route retrieval execution can be carried out correctiy.
It will be obvious to those skilled in the art that various changes may be made in the above-described embodiments of the present invention. However, the scope of the present invention should be determined by the following claims.
Claims (7)
- WHAT IS CLAIMED: 1. A code recognition apparatus for recognizing a code among codes of a specific kind, each of the codes satisfying a predetermined rule of the specific kind and being a character string consisting of a predetermined first portion and a predetermined second portion, the second portion being also a character string, the code recognition apparatus comprising: a Sound collecting means for collecting a sound in a surrounding area and transforming the collected sound to thereby generate a sound signal: a first recognizing means for recognizing a first portion of the codes of the specific kind based on the sound signal generated by the sound collecting means; an enumerating means for enumerating a whole of character strings of second portions, wherein the predetermined rule is satisfied when each of the whole of character strings is united with the first portion recognized by the first recognizing means; a dictionary preparing means for preparing a speech recognition dictionary in which the whole of the character strings of the second portions enumerated by the enumerating means and sound data elements are respectively associated with each other; a second recognizing means for recognizing, as a character string of a second portion, a character string corresponding to a sound data element most matching with a sound signal generated by the sound collecting means, from among the sound data elements included in the speech recognition dictionary after the dictionary preparing means prepares the speech recognition dictionary; and an extracting means for extracting a predetermined sub portion, which is included in the second portion, from the character string recognized by the second recognizing means.
- 2. The code recognition apparatus according to claim 1, further comprising: a storing means for storing a database containing a data element corresponding * to each of the codes of the specific kind, the data element being associated with each combination consisting of (i) a character string of a first portion to be recognized by the first recognizing means and (ii) a character of a sub portion in a second portion to be extracted by the extracting means; and an acquiring means for acquiring, from the database, a data element associated with each combination consisting of (i) the character string of the first portion recognized by the first recognizing means and (ii) the character of the sub portion in the second portion extracted by the extracting means.
- 3. The code recognition apparatus according to claim 2, wherein: each of the codes of the specific kind is a postal code number which consists of a combination of a character string of a preceding portion and a character string of a succeeding portion; the first portion is the preceding portion of the postal code number; the second portion is the succeeding portion of the postal code number; and the sub portion in the second portion is a part of the succeeding portion of the postal code number, the sub portion in the second portion succeeding the preceding portion of the postal code number.
- 4. The code recognition apparatus according to claim 3, wherein: the first portion consists oan optional combination of a numeral character and an alphabetical character, the second portion consists of a single numeral character and two alphabetical characters succeeding the single numeral character; the sub portion is a single numeral character included in the second portion; the enumerating means (i) retrieves a whole of character strings, each of which includes the first portion recognized by the first recognizing means as a first portion of a postal code number, from a group of character strings respectively consisting of first portions and sub portions, the group of the character stnngs being registered in the database as combinations of the first portions and the sub portions, (ii) designates a whole of numeral characters consisting sub portions of the retrieved whole of character strings; (iii) enumerates, with respect to each of the designated whole of numeral characters, a whole of combinations of two alphabetical characters succeeding the each of the designated whole of numeral characters; and (iv) forms character strings constituting second portions by combining the each of the designated whole of numeral characters and the enumerated whole of combinations of two alphabetical characters.
- 5. A route retrieval apparatus comprising; the code recognition apparatus according to claim 3 or claim 4; a retrieval means for retrieving a route between spots, wherein: the database includes a spot data element with respect to each of the character stnngs consisting of (i) the preceding portions and (ii) sub portions of the second portions succeeding the preceding portions; the acquiring means acquires a spot data element from the database, based on a character string of the preceding portion recognized by the first recognizing means and a sub portion of the succeeding portion extracted by the extracting means, the sub portion directly succeeding the preceding portion; and the retrieving means designates, as a start point or an end point of a route to be retrieved, a spot indicated by the spot data element acquired by the acquiring means.
- 6. A code recognition apparatus substantially as described herein with reference to the accompanying drawings.
- 7. A route retrieval apparatus substantially as described herein with reference to the accompanying drawings.
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
JP2008232445A JP4645708B2 (en) | 2008-09-10 | 2008-09-10 | Code recognition device and route search device |
Publications (3)
Publication Number | Publication Date |
---|---|
GB0915921D0 GB0915921D0 (en) | 2009-10-14 |
GB2463371A true GB2463371A (en) | 2010-03-17 |
GB2463371B GB2463371B (en) | 2012-05-30 |
Family
ID=41228158
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
GB0915921.1A Expired - Fee Related GB2463371B (en) | 2008-09-10 | 2009-09-10 | Code recognition apparatus and route retrieval apparatus |
Country Status (2)
Country | Link |
---|---|
JP (1) | JP4645708B2 (en) |
GB (1) | GB2463371B (en) |
Families Citing this family (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JP7400112B2 (en) | 2020-01-17 | 2023-12-18 | グーグル エルエルシー | Biasing alphanumeric strings for automatic speech recognition |
Citations (7)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US5635953A (en) * | 1991-07-11 | 1997-06-03 | Mitsubishi Denki Kabushiki Kaisha | Map displaying apparatus |
WO2000005710A1 (en) * | 1998-07-21 | 2000-02-03 | British Telecommunications Public Limited Company | Speech recognition |
JP2002073074A (en) * | 2000-08-31 | 2002-03-12 | Pioneer Electronic Corp | Method and device for recognizing numerical string in voice |
US20030101045A1 (en) * | 2001-11-29 | 2003-05-29 | Peter Moffatt | Method and apparatus for playing recordings of spoken alphanumeric characters |
JP2007011358A (en) * | 2005-06-28 | 2007-01-18 | Avaya Technology Corp | Speech recognition assisted autocompletion of composite character |
JP2007178692A (en) * | 2005-12-27 | 2007-07-12 | Canon Inc | Character input device and method |
EP2096412A2 (en) * | 2008-02-29 | 2009-09-02 | Navigon AG | Method for operating a navigation system |
Family Cites Families (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JPH04271397A (en) * | 1991-02-27 | 1992-09-28 | Sanyo Electric Co Ltd | Voice recognizer |
JPH07120296A (en) * | 1993-10-22 | 1995-05-12 | Yamatake Honeywell Co Ltd | Liquid level meter |
JP3865149B2 (en) * | 1995-08-22 | 2007-01-10 | 株式会社リコー | Speech recognition apparatus and method, dictionary creation apparatus, and information storage medium |
JP2006003142A (en) * | 2004-06-16 | 2006-01-05 | Matsushita Electric Ind Co Ltd | Number input device and navigation system using the same |
JP4758758B2 (en) * | 2005-12-26 | 2011-08-31 | 日本放送協会 | Dictionary creation device and dictionary creation program |
-
2008
- 2008-09-10 JP JP2008232445A patent/JP4645708B2/en not_active Expired - Fee Related
-
2009
- 2009-09-10 GB GB0915921.1A patent/GB2463371B/en not_active Expired - Fee Related
Patent Citations (7)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US5635953A (en) * | 1991-07-11 | 1997-06-03 | Mitsubishi Denki Kabushiki Kaisha | Map displaying apparatus |
WO2000005710A1 (en) * | 1998-07-21 | 2000-02-03 | British Telecommunications Public Limited Company | Speech recognition |
JP2002073074A (en) * | 2000-08-31 | 2002-03-12 | Pioneer Electronic Corp | Method and device for recognizing numerical string in voice |
US20030101045A1 (en) * | 2001-11-29 | 2003-05-29 | Peter Moffatt | Method and apparatus for playing recordings of spoken alphanumeric characters |
JP2007011358A (en) * | 2005-06-28 | 2007-01-18 | Avaya Technology Corp | Speech recognition assisted autocompletion of composite character |
JP2007178692A (en) * | 2005-12-27 | 2007-07-12 | Canon Inc | Character input device and method |
EP2096412A2 (en) * | 2008-02-29 | 2009-09-02 | Navigon AG | Method for operating a navigation system |
Also Published As
Publication number | Publication date |
---|---|
JP4645708B2 (en) | 2011-03-09 |
JP2010066493A (en) | 2010-03-25 |
GB0915921D0 (en) | 2009-10-14 |
GB2463371B (en) | 2012-05-30 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
US7310602B2 (en) | Navigation apparatus | |
US9105267B2 (en) | Speech recognition apparatus | |
EP1505571A1 (en) | Car navigation system and speech recognizing device thereof | |
EP2226793A2 (en) | Speech recognition system and data updating method | |
JPH10253381A (en) | Voice synthesizer | |
JP4466379B2 (en) | In-vehicle speech recognition device | |
US20050171685A1 (en) | Navigation apparatus, navigation system, and navigation method | |
EP2660562A1 (en) | Route Guidance Apparatus and Method with Voice Recognition | |
CN101276585A (en) | Multilingual non-native speech recognition | |
JPH0764480A (en) | Voice recognition device for on-vehicle processing information | |
JP5455355B2 (en) | Speech recognition apparatus and program | |
JP2005275228A (en) | Navigation system | |
WO2014199428A1 (en) | Candidate announcement device, candidate announcement method, and program for candidate announcement | |
JP2000338993A (en) | Voice recognition device and navigation system using this device | |
EP0827124A2 (en) | Vehicle navigation system with city name selection accelerator and medium for storage of programs thereof | |
GB2463371A (en) | Retrieving route information using speech recognition and spoken postal codes | |
JP4381632B2 (en) | Navigation system and its destination input method | |
JP2011232668A (en) | Navigation device with voice recognition function and detection result presentation method thereof | |
JPH0926799A (en) | Speech recognition device | |
WO2006028171A1 (en) | Data presentation device, data presentation method, data presentation program, and recording medium containing the program | |
JP2000122685A (en) | Navigation system | |
US10832675B2 (en) | Speech recognition system with interactive spelling function | |
JP2007280104A (en) | Information processor, information processing method, information processing program, and computer readable recording medium | |
JP3579971B2 (en) | In-vehicle map display device | |
JP2008157885A (en) | Information guide device, navigation device, information guide method, navigation method, information guide program, navigation program, and recording medium |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
746 | Register noted 'licences of right' (sect. 46/1977) |
Effective date: 20130228 |
|
PCNP | Patent ceased through non-payment of renewal fee |
Effective date: 20130910 |