JPH10198549A - Information processor, navigation device, speech inputting method and recording medium - Google Patents

Abstract

PROBLEM TO BE SOLVED: To improve the recognition ratio of an operation instruction with speech at low costs in a navigation device in which the input of the operation instruction with speech can be attained. SOLUTION: When a user utters 'menu', a menu matrix M is screen displayed. The menu matrix M is composed of plural lines and plural columns. A line number (a) is assigned to each line, and a column number (b) is assigned to each column, and a command such as 'enlargement', 'detail', 'departure', and 'destination' is character-displayed at a position indicated by the line number (a) and the column number (b). When a word uttered by the user is, for example, 'two-one', the word is voice recognized, and it is judged that the command of 'detail' in the second line/first column is designated, and a map displayed at that point is switched to a detail map. Since the designation of the command is operated by digits easy in speech recognition, the recognition ratio of the designated command can be increased even under the environment with high noises.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an information processing apparatus and a navigation apparatus which enable an operation instruction by voice.
And a voice input method and a recording medium.

[0002]

2. Description of the Related Art In recent years, a navigation device that guides a driver to a destination while displaying a map near the position of a vehicle has become popular. In such a navigation device,
When selecting commands for operating the displayed menu screen, perform key operations while looking at the menu screen,
When the menu screen is displayed on the touch panel, command selection is performed by directly touching a command portion. Also, in recent ones, in order to enable operation while traveling and to improve operability, commands recognized by the user using voice recognition technology are recognized,
Some have enabled command input by voice.

[0003]

However, since the navigation device is used in an environment where there is a lot of noise such as the sound of audio equipment in the vehicle compartment and the noise outside the vehicle compartment, the navigation device is not affected by those noises. ,
It is difficult to correctly recognize a command issued by a user, and there is a problem that increasing the command recognition rate increases the cost required for voice recognition and increases the size of the device.

SUMMARY OF THE INVENTION The present invention has been made in view of the above-mentioned conventional problems, and an information processing apparatus and a navigation apparatus capable of improving the recognition rate of operation instructions by voice at low cost, and a voice input method and recording. The purpose is to provide a medium.

[0005]

According to an embodiment of the present invention, there is provided an information processing apparatus, comprising: a display for displaying a menu screen showing a correspondence between a command and designation information for designating the command; Means, voice input means for inputting voice, voice recognition means for recognizing the specified information by voice input by the voice input means, and a command specified by the specified information recognized by the voice recognition means And control means for executing processing according to the above.

In this configuration, when any of the designated information displayed on the menu screen is recognized by voice, a process corresponding to a command designated by the designated information is executed. Here, in speech recognition, it is sufficient to recognize not the command itself but the designation information for designating the command. Therefore, as the designation information necessary for the execution of the command, words that can be easily recognized as speech, for example, numbers and characters ( One word).

In another information processing apparatus according to the present invention, a voice input means for inputting voice, a command based on the voice input by the voice input means, and designation information for designating the command are provided. Voice recognition means for voice recognition;
Control means for determining a command specified by the specification information recognized by the voice recognition means and executing a process corresponding to the determined command.

In this configuration, when the designated information is recognized by speech, a process corresponding to the command designated by the designated information is executed. Here, in speech recognition, it is sufficient to recognize not the command itself but the designation information for designating the command. Therefore, as the designation information necessary for the execution of the command, words that can be easily recognized as speech, for example, numbers and characters ( One word). In addition, by determining that the same designation information indicates different commands under different circumstances, it becomes possible to execute a plurality of commands with the same designation information.

Further, in the navigation device of the present invention, a display means for displaying a map near a self-position based on the map data read from the map data storage means, a voice input means for inputting voice, and the voice input means Voice recognition means for recognizing the specified information by voice input by the means, and control means for determining a command specified by the specified information recognized by the voice recognition means and executing a process corresponding to the determined command And was provided. In such a configuration, when the designation information for designating the command is recognized, a process corresponding to the command designated by the designation information is executed. Therefore, it is possible to perform an operation by voice and to use, for example, a number or a character (one word) as designation information necessary for executing the command.

Further, in the method of the present invention, in the method of inputting an operation instruction by voice, voice-inputted specification information for specifying a command is recognized by voice and specified by the voice-recognized specification information. The processing according to the command was executed. According to such a method, as described above, in speech recognition, it is sufficient to recognize not the command itself but the designation information for designating the command. Therefore, as the designation information necessary for executing the command, for example, a number or a character ( One word).

The recording medium of the present invention is a recording medium on which a program which is read and executed by a computer is recorded, and which responds to a command indicated by designation information for designating a command, which is recognized by voice. A program for executing the process was included. If the computer is operated using the recording information of the recording medium, when the designation information for designating the command is recognized by speech, a process corresponding to the command indicated by the designation information is executed. Therefore, in a device equipped with a computer, a voice operation can be performed, and, for example, a number or a character (one word) can be used as a necessary operation word necessary for executing a command.

[0012]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS One embodiment of the present invention will be described below with reference to the drawings. FIG. 1 is a block diagram showing a navigation device according to the present invention. This navigation device has general functions such as a route calculation function, a route guidance function, and a point registration function, and a command input function by voice, and has a GPS antenna 1 as a reception function,
C in L1 band from satellite received by GPS antenna 1
And a GPS block 2 for demodulating and decoding the / A code to determine the latitude and longitude of the current location. The GPS block 2 is connected to the graphic block 3
The latitude / longitude information determined by the block 2 is input to the graphic block 3 and taken into the CPU 4. CPU block G / A
5, a timer unit 6, a ROM 7, a RAM 8, and a graphic controller 9 for measuring the timing of the processing operation and the like of the CPU 4 are connected to the CPU 4, respectively, and a video RAM 10 is connected to the graphic controller 9. In the ROM 7, a program for causing the CPU 4 to control the entire navigation device using the RAM 8 as a working area, operation words (numbers in the present embodiment) as designation information for designating a plurality of commands, and Various types of data such as a command correspondence table (not shown) including a plurality of commands corresponding to each are stored.

The graphic block 3 is connected to a CD block 11 and a monitor 12 as display means of the present invention. The CD block 11 is used to display data stored in the map CD-ROM 13, that is, map / road information such as road data and place name data, traffic regulation data used for automatic route calculation, and the like, and display the traffic information and the like on a map. Read and output the road link data used for. CD
Various data output from the block 11 are sent to the CPU 8 via the RAM 8 together with the data read from the ROM 7.
It is sent to the CPU 4 via the peripheral circuit G / A5. Further, the data read from the CD-ROM 13 is sent to the monitor 12 via the video RAM 10 by the graphic controller 9, and a map showing the position of the vehicle and the surrounding area is displayed on the monitor 12. The graphic block 3 includes a D / A converter 14.
A speaker 16 is connected via an amplifier 15 and a guidance sound composed of a synthesized sound is emitted from the speaker 16 as needed at the time of route guidance or the like.

Further, the graphic block 3 is connected to a key block 17 having various operation keys (not shown) for operating the navigation device, and a voice recognition block 18 as voice recognition means of the present invention. The voice recognition device 18 includes a voice input device 1 having a microphone.
9 is connected. The voice uttered by the user is input to the voice recognition block 18 via the voice input device 19, and after being recognized as an operation word by the voice recognition block 18, C
Sent to PU4. In the present embodiment, the voice recognition block 18 enables voice recognition of only numbers and voices of “menu”.

Next, the operation of the navigation device having the above configuration will be described. This flowchart shows the operation when the voice input mode is selected. When the voice input mode is selected, the navigation device starts operation and waits for voice input (S
1). Then, in this state, when the user utters a voice (word) and there is a voice input to the voice input device 19 (S
If the speech recognition block 18 starts speech recognition processing and speech recognition is performed, the recognition result is sent to the CPU 4 as an operation word (S2). It is determined whether or not an input has been made (S3). If the result of this determination is NO, the process returns to step S1, and if YES, the menu matrix M shown in FIG. 3 is displayed on the screen of the monitor 12 at that time, for example, a map screen. (S
4). The menu matrix M is composed of a plurality of rows and a plurality of columns, and each row has a row number a of “1”, “2”, “3”, “4”,.
, And each column has "1", "2",
The column number b of “3” is assigned. Then, commands prepared in the navigation device are displayed in characters at positions indicated by the row numbers a and the column numbers b. In the present embodiment, commands such as “enlarge”, “details”, “departure place”, “destination”,. At the line number "4" / column number "3", an "end" command for terminating the display of the menu matrix M is displayed in characters. Although omitted in FIG. 3, other commands are also displayed in characters at positions indicated by other row numbers and column numbers.

After the menu matrix M is displayed in step S4, the process waits for voice input again (S5). If there is no voice input, the process returns to step S4. Recognition processing is started, and when voice recognition is performed, the recognition result is sent to the CPU 4 as an operation word (S6), and the CPU 4 determines whether or not the line number a of the menu matrix M in FIG. It is determined whether two numbers within the maximum value have been consecutively input as operation words in the order of rows and columns (S7). That is, it is determined whether or not the user has issued an operation word for designating any command in the menu matrix M. Here, if the voice recognized in step S6 is, for example, "Ichi Yong" or "San Go" and the determination result in step S7 is NO, the process returns to step S4, for example, "Ichi San" or "San Go". If the determination result of step S7 is YES, the process proceeds to the next step.

In the following step S8, based on the operation word recognized in step S6, it is determined which command on the menu matrix M the user has selected by referring to the aforementioned command correspondence table stored in the ROM 7. I do. For example, when the recognized voice is “2/1”, it is determined that the selected command is a command of 2 rows / 1 column, that is, “detailed” in the present embodiment. Thereafter, a process corresponding to the determined command is executed (S9), and the process ends. Note that, in step S7, Y
The display of the menu matrix M is terminated when it is determined as ES. When the user utters, for example, the word "Ni-ichi", the map displayed on the monitor 12 is switched to the detailed map.

As described above, in the navigation apparatus, it is possible to input a command by voice only by recognizing only two numbers emitted by the user, that is, words for which voice recognition is easy. Therefore, even in an environment where there is much noise such as noise outside the vehicle cabin, the command specified by the user can be correctly recognized without being affected by such noise. In addition, since the voice recognition does not require a complicated recognition operation, the cost required for the voice recognition block 18 is low. In addition, the operation word required for executing the command is not limited to a number, and for example, a simply encoded operation word having no meaning as a command by itself such as one letter of the Japanese syllabary or one letter of the alphabet is used. You may. Even in that case, the same effect can be obtained.

Further, in the navigation apparatus, since one command is determined by using two numbers input by voice as elements of a matrix, more commands can be obtained by recognizing only a few types of numbers and the like (words). Command can be executed. Therefore, even when the number of commands is large, the specified command can be recognized at a high recognition rate. Moreover, since the number corresponding to the desired command can be known from the menu matrix M displayed on the monitor 12, it is not necessary to remember the correspondence between the encoded numbers and the like (words) and the commands, and the usability is improved. Good.

In the present embodiment, a case where one menu matrix M is displayed has been described. However, a plurality of menu matrices are prepared in advance, and for example, a voice input such as "@menu" is firstly performed, A menu matrix desired by the user may be selectively displayed according to a key operation in the key block 17. In that case, more commands can be executed only by recognizing fewer types of numbers and the like (words).

The command selected at the time of command determination in step S8 or at the end of the above-described operation is stored, and the next time the menu matrix M is displayed on the display screen of the monitor 12, the previously specified and executed command is executed. While performing the mark king process on the character indicating the command, at the time, for example, "left (hidden)",
When there is a voice input of a simple word such as “right (Migi)”, “up (we)”, “down (sita)”, etc., it is adjacent to the left, right, up and down of the marked character in the menu matrix M. A process for assuming that the command has been selected may be performed. In that case, the operability for specifying the command is improved.

On the other hand, in the present embodiment, the description has been given of the case where one command is designated by two numbers or the like (words).
May be used to specify one command. In other words, numbers and the like (words) and commands may be set in a one-to-one relationship, and commands may be input by a menu having one column or row unlike the menu matrix M. In this case, If the number of commands is small, the menu display may be omitted.

Also, a number or the like (word) and a command may be set in a one-to-many relationship. This is achieved by, for example, preparing a plurality of menu matrices M and command correspondence tables in which commands indicated by operation words such as numbers (words) are different from each other, and selectively inputting commands using these. Can do. In particular,
The voice recognized in step S3 is "menu"
In the subsequent step S4, a matrix menu in which only commands valid at that time and commands that are likely to be used at that time are displayed based on the operation contents of the navigation device is selectively displayed. Thereafter, if the input voice is a valid number or the like (word) in the menu, an operation of determining and executing a corresponding command in the menu matrix displayed at that time may be performed. . In such cases,
More commands can be executed only by recognizing fewer types of numbers and the like (words). Further, such a command input method can be performed using the above-described menu having one column or one row. In the present embodiment, since a command is input using one menu matrix M, when a number or the like (word), which is an operation word, is input, it is determined which command corresponds to it. No processing is required. Therefore, the processing at the time of command input is simplified as compared with the case of using a plurality of menu matrices M (or a menu having one column or row).

In the present embodiment, CUP4
Has been described in the case where the control program is recorded in the ROM 7 in advance. However, a part or all of the control program is recorded on another recording medium such as the CD-ROM 13 mounted on the CD book 11, and the control program is started. With RAM
Alternatively, the configuration may be such that the data is read into the memory 12 or the like.
In particular, in a navigation apparatus in which new functions tend to increase and an increase in operation commands is expected, it is convenient because the new functions can be easily handled.
Further, it is convenient when the audio information processing apparatus of the present invention is realized by a general personal computer and other apparatuses including a computer and the like.

[0025]

As described above, in the present invention,
When the designation information for designating the command is recognized by speech, a process corresponding to the command designated by the designation information is executed. As a result, as the designation information necessary for executing the command, for example, a number or a character ( (One word) for easy speech recognition. Therefore, by using words that are easy to recognize by voice as designation information necessary for executing the command, the recognition rate of operation instructions by voice can be improved at low cost.

In the information processing apparatus according to the present invention, if the control means determines a command indicated by the operation word and executes processing in accordance with the determined command,
By determining that the same designation information indicates different commands under different situations, it becomes possible to execute a plurality of commands with the same designation information. Therefore, in addition to the above effects, many commands can be executed with a small number of operation words, and even when the number of commands is large, a high recognition rate for operation instructions can be secured at low cost. Is obtained.

Further, if the control means determines a command using the two pieces of designation information recognized by the speech recognition means as elements of a matrix, a greater number of commands can be indicated with less designation information. Therefore, even when the number of commands is large, the specified command can be recognized with a higher recognition rate at low cost.

Further, in the apparatus provided with a display means for displaying a menu screen indicating the correspondence between the designated information and the command, the user knows the designated information corresponding to the desired command while looking at the displayed menu screen. be able to. Therefore, it is not necessary to remember the correspondence between the command and the designation information for designating the command, and the use environment is improved.

Further, if the voice recognition means is capable of performing voice recognition of a command related to display of a menu screen, operability is further improved.

[0030]

[Brief description of the drawings]

FIG. 1 is a block diagram of a navigation device showing one embodiment of the present invention.

FIG. 2 is a flowchart showing an operation when a voice input mode is set in the navigation device.

FIG. 3 is a diagram showing a menu matrix displayed on a monitor in a voice input mode.

[Description of Signs] 2 GPS block 4 CPU 7 ROM 8 RAM 11 CD block 12 Monitor 18 Voice recognition block 19 Voice input block

Claims (8)

[Claims] 1. A display means for displaying a menu screen showing a correspondence between a command and designation information for designating the command, a voice input means for inputting voice, and a voice input by the voice input means. An information processing apparatus comprising: voice recognition means for recognizing the specified information by voice; and control means for executing a process corresponding to a command specified by the specified information recognized by the voice recognition means. 2. A voice input means for inputting voice, voice recognition means for voice-recognizing a command and designation information for specifying the command by voice input by the voice input means, An information processing apparatus comprising: a control unit configured to determine a command specified by the recognized specification information and execute a process corresponding to the determined command. 3. The information processing apparatus according to claim 2, wherein the control unit determines the command using the two pieces of designated information recognized by the voice recognition unit as elements of a matrix. 4. The information processing apparatus according to claim 2, further comprising display means for displaying a menu screen showing a correspondence between the designation information and the command. 5. The information processing apparatus according to claim 1, wherein the voice recognition unit is capable of performing voice recognition of a command related to the display of the menu screen. 6. A display means for displaying a map near the self-position based on the map data read from the map data storage means; a voice input means for inputting voice; Voice recognition means for recognizing specified information by voice, and control means for determining a command specified by the specified information recognized by the voice recognition means and executing processing in accordance with the determined command. Navigation device. 7. A method for inputting an operation instruction by voice, wherein voice recognition of voice-inputted specification information for specifying a command is performed, and processing corresponding to the command specified by the voice-recognized specification information is performed. A voice input method characterized by the following. 8. A recording medium storing a program read and executed by a computer, the program including a program for executing a process corresponding to a command indicated by designation information for designating a command, which is recognized by voice. Recording medium characterized by the above-mentioned.