JP2000029585A - Voice command recognition image processing device - Google Patents

Abstract

(57) [Summary] [PROBLEMS] A memory or the like for voice dictionary data to be registered in advance when used by an unspecified number of people in a copier or a multifunction machine such as a facsimile or a printer that operates by recognizing voice input commands Means for mitigating an excessive increase in the capacity of the storage medium. For this reason, when registering reference data in dictionary data, the reference data is registered while actually pressing a button or the like of an operation unit 101, and the next time the same command is used, it can be operated only by voice, and When performing registration, pattern matching is performed, and if the same or similar command or similar voice is input as a registered command, a warning is issued to the user, and dictionary data having a high similarity is It is configured to recognize more accurately with other judgment factors such as fingerprints and palm prints.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a multifunction machine such as a copier and a facsimile or a printer, and more particularly to a multifunction machine using a voice for a user interface for inputting a command.

[0002]

2. Description of the Related Art In recent years, devices for inputting commands using voice recognition, such as those found in personal computers and car navigation systems, have become widespread. However, in conventional copiers and multifunction machines such as facsimile machines and printers. , None of the above-mentioned functions.

[0003] To perform this type of speech recognition, "pattern matching" such as voiceprint and voice delimitation and pitch of sound is performed. The pattern matching means that a command to be used in a storage medium such as a memory and voice data when the command is uttered are registered as a dictionary in one-to-one correspondence, and the input voice data is registered. Is calculated as the degree of similarity with each command
The command with the largest value is selected and executed.

[0004]

However, in the above-described copying machine and multifunction machines such as facsimile machines and printers,
There are the following problems in inputting commands by voice recognition as described above.

That is, when a command is input by voice recognition, all commands to be used must be registered in a dictionary in advance. In the current command registration method, all the commands to be used must be read once and registered in correspondence with each key operation.

[0006] In a car navigation system, a personal computer, or the like, in which the number of users is limited to some extent as in the past, the data to be registered is generally personal data.
It was enough to just input human voice data.

However, in the case of a machine that is used by an unspecified number of people, such as a copying machine, it is necessary to register voice dictionary data of many people who may use the machine in advance. A storage medium such as a memory used for storage also needs to have a relatively large capacity.

[0008] In addition, the commands to be registered also have different frequencies of use, and even if there are commands that are not used at all depending on the person, it is necessary to register all of them. Requires the use of unnecessary storage medium capacity.

SUMMARY OF THE INVENTION The present invention has been made in view of the above situation, and has as its object to provide a voice command recognition image processing apparatus capable of solving these problems.

[0010]

Therefore, in the present invention, the above object is achieved by providing a voice command recognition image processing apparatus as described in any of the following items (1) to (6). What we are trying to achieve.

(1) Operation means for inputting commands such as copy operation and facsimile transmission by switches and buttons, numeric keys, and a liquid crystal touch panel, and a processing device such as a CPU. A driver means for setting image processing, setting the number of prints, etc., and controlling operations of an image input means, an image processing means, an image output means, and the like. Image input means for converting the image data into digital data and outputting the digital data to an image processing unit;
The digital image data input from the image input unit is processed by a hardware circuit such as an SIC or the like by software processing using the CPU or the like in accordance with an image processing setting from the driver unit. Image processing means for outputting to the image output means; and image output means for outputting the image data processed by the image processing means by a printer such as a laser printer or an ink jet printer. LCD monitor,
Alternatively, the image output unit including a unit that performs display on the liquid crystal touch panel used in the operation unit, performs image data output on a network, performs facsimile transmission via a modem, and the like, and uses a microphone and the like. And
Speech input means for inputting speech and sending it to the speech recognition means as the digital data, and speech data from the speech input means for registration in dictionary data, pattern matching of speech, conversion to commands, etc. A voice command recognition image processing apparatus comprising: voice recognition means; and operating by recognizing the command by voice.

(2) The voice recognition means includes voice data input means for inputting digital voice data from the voice input means, operation section information input means for inputting command information input from the operation means, A dictionary data storage unit that stores the operation unit information and the voice data in a storage medium such as a memory or a hard disk, and data from the voice data input unit, and data from the dictionary data storage unit. Pattern matching means for performing pattern matching with voice dictionary data and outputting a degree of similarity with respect to each command; driver information input means for inputting the dictionary information registration and normal copy operation information from the driver means; In the case of normal copy, receiving information from the matching means and information from the command information input means Command processing means to determine which command the input voice command corresponds to, and command information to be sent to command output means; and command information from the command processing means. The voice command recognition image processing apparatus according to (1), further comprising command output means for sending to the mode display portion.

(3) The voice command recognition image processing apparatus, when registering data in the dictionary data storage means,
The voice command recognition image processing apparatus according to any one of the above (1) and (2), wherein the voice data is registered while actually operating.

(4) When registering the voice data, the voice command recognition image processing apparatus uses the fact that the same data as the already registered data is to be registered in the dictionary data storage means. The voice command recognition image processing apparatus according to any one of (1), (2), and (3), wherein the voice command recognition image processing apparatus is configured to notify the user.

(5) The voice command recognition image processing apparatus notifies the user of the unregistered command if the unregistered command is set by the voice when the voice command is set. The voice command recognition image processing apparatus according to any one of the above items (1) to (4), characterized in that:

(6) The voice command recognition image processing apparatus notifies the user of the command when the voice input command is similar to a plurality of commands when performing the setting by the voice. 7. The voice command recognition image processing apparatus according to any one of the above items (1) to (5), wherein the voice command recognition image processing apparatus prompts the user to make a selection.

[0017]

According to the configuration of the present invention as described above, it is possible to input a new voice command which solves the above-mentioned problems. 1) When registering data to be referred to the dictionary data, the registration is performed by actually registering the data while pressing a button or the like of the operation unit. When the same command is used next time, the operation can be performed only by voice. By doing so, it is not necessary for each user to register all commands by taking a time in advance, and it is possible to register only commands frequently used by each person who uses the machine. This can reduce the capacity of the storage medium used for the dictionary.

2) Also, when registering data, pattern matching is performed, and when the same command as the already registered command is input, or when a similar command or similar voice is input, display or display is performed. By notifying the user with a warning sound or the like, it is possible to confirm whether or not the command to be registered is to be properly input.

3) Also, when the user recognizes by voice, if there are a plurality of registered dictionary data having a high similarity to some extent, other dictionary factors such as fingerprints and palm prints are used. Thus, by performing recognition more accurately, it is possible to more accurately identify an individual than to perform recognition of a user only by voice.

[0020]

DESCRIPTION OF THE PREFERRED EMBODIMENTS Embodiments of the present invention will be described below.
A plurality of embodiments will be described in detail with reference to the drawings.

[0021]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS (First Embodiment) A specific first embodiment of an image processing apparatus according to the present invention will be described using an example of a copying machine. FIG. 1 shows a configuration block diagram of a copying machine according to the first embodiment (FIG. 1 is also used in the second to fifth embodiments described later).

An operation unit 101 includes switches, buttons, ten keys, and the like as shown in FIG.
A command such as copy operation or facsimile transmission is input by a liquid crystal touch panel or the like. The description of the individual components in FIG.

Reference numeral 102 denotes a driver unit which uses a processing device such as a CPU to set image processing and the number of prints in accordance with a command from the operation unit 101. The scanner unit 103, the image processing unit 104, and the printer unit The control of the operation such as 105 is performed.

Reference numeral 103 denotes an image input unit which uses a CCD or the like, converts image data of a document into digital data under the control of a driver unit 102, and outputs the digital data to an image processing unit 104.

Reference numeral 104 denotes an image processing unit which converts digital image data input from the image input unit 103 into image processing from the driver unit 102 by a hardware circuit such as an ASIC or software processing using a CPU or the like. Is performed in accordance with the setting of (1), and output to the image output unit 105.

Reference numeral 105 denotes an image output unit;
The processed image data is output by a printer such as a laser printer or an ink jet printer, or is output to a CRT monitor, a liquid crystal monitor, a touch panel used for the operation unit 101, or the like. In addition, output to the network as image data and facsimile transmission via a modem are also included in this part.

Reference numeral 106 denotes a voice input unit, which inputs a voice using a microphone or the like and sends the digital data to a voice recognition unit.

Reference numeral 107 denotes a voice recognition unit, and a voice input unit 106
, And performs registration in dictionary data, pattern matching of voice, conversion into commands, and the like. The converted command is sent to the driver unit 102, similarly to the command from the operation unit 101.

Next, the speech recognition unit 10 which is a feature of the present invention is described.
7 will be described in detail with reference to FIG. 2 which shows a block diagram of the internal processing configuration (this drawing is also used in FIG. 1 and in the second to fifth embodiments described later).

Reference numeral 201 denotes an audio data input unit for inputting digital audio data from the audio input unit 201.

Reference numeral 202 denotes an operation unit information input unit, and the operation unit 10
The command information input from step 1 is input.

Reference numeral 203 denotes a dictionary data storage unit which stores operation unit information and voice data in a storage medium such as a memory or a hard disk in association with each other.

Reference numeral 204 denotes a pattern matching unit which stores data from the voice data input unit 201 and a dictionary data storage unit 2
The pattern matching is performed with the voice dictionary data from No. 03, and the similarity for each command is output.

For example, when the input voice data is “ichima”, the similarity to “ichima” of the voice dictionary data is 100, the similarity to “hachimai” is 80,
The similarity with “sanmai” is output as 30, the similarity with “copy” is 5, and the like.

A driver information input unit 205 inputs information such as dictionary information registration and a normal copy operation from the driver unit 102.

A command processing unit 206 includes information from the pattern matching unit 204 and a command information input unit 205.
In the case of a normal copy, it is determined which command the input voice command corresponds to,
The command is sent to the command output unit 207. When registering the dictionary information, processing such as sending some of the commands having high similarity to the command output unit 207 is performed.

A command output unit 207 sends command information from the command processing unit 206 to the mode display unit of the driver unit 102 or the operation unit 101 in the case of normal copying. When registering dictionary information, a signal for generating a similar command or a warning sound is sent to the operation unit 101.

Next, the contents of the process will be described with reference to a specific example: When registering a voice command, the user speaks while actually pressing a predetermined key. At this time, the audio data input unit 2
From 01, the voice data at that time is input, and from the operation unit information input unit 202, information on the key pressed at that time is input.

For example, when performing "one monochrome copy", "copy" is uttered while pressing the "copy" key in FIG. Then, from the operation unit 101 to the driver unit 102
Is transmitted to the operation unit information input unit 202 of the voice recognition unit 107 in the same manner.

The voice data from the voice input unit 106 is input to the voice data input unit 201 and is input to the dictionary data storage unit 203 together with command information. The dictionary data storage unit 203 stores the input voice data in response to the command “copy”.

Hereinafter, similarly, the command "voice" and voice data and the command "single" and voice data are registered and the setting is performed at the same time.

While holding down the start key,
As a result, the voice command "start" is registered and the copy output can be performed.

Through the above operations, voice commands such as "copy", "black and white", "one sheet", and "start" are registered while actually performing copying.

The above operation sequence flowchart at the time of registering a voice command corresponds to a portion indicated by a solid line arrow in FIG. Voice input (step S301) is performed by the voice data input unit 201 in FIG. 2, and the status of the operation unit at that time is detected by the operation unit status input unit 202, and it is determined whether or not key input is being performed simultaneously. (Step S302).

When the key input is performed at the same time, the operation enters the voice data registration mode (step S303). The pattern matching of the voice data is performed by the pattern matching unit 204 (step S304). If there is no similar command (step S305), the input voice data is registered as a command in the dictionary data storage unit 203 (step S306). The command is executed (step S307).

It should be noted that the detailed description of each step indicated by a dotted-line arrow indicating NO in each of the above steps is as shown in the drawing, and a repeated description of the sequential details will be omitted.

The command registered by the above method can be used only by voice without performing key operation.

That is, if one wants to make one black and white copy,
Speaking "copy" switches to "copy" mode, and uttering "black and white" switches to "black and white" mode. Thereafter, by saying “one sheet” and “start”, a monochrome copying operation can be performed.

The operation sequence flowchart at the time of executing the voice command at this time corresponds to the portion indicated by the solid line arrow in FIG. Voice input (step S401) is performed by the voice data input unit 201 in FIG. 2, and the status of the operation unit 101 at that time is detected by the operation unit status input unit 202, and it is determined whether or not key input is being performed simultaneously. (Step S40
2).

If the key input is not performed at the same time, the mode is the voice data execution mode (step S403). The voice data is subjected to pattern matching by the pattern matching unit 204 (step S404). If there is only one command similar to the registered command (steps S405 and S406), the input voice data is used as a command. Execute (step S407).

Regarding the recognition of the voice command at this time,
The voice command is recognized in detail by pattern matching between voice data from the voice data input unit 201 in FIG. 2 and voice data stored in the dictionary data storage unit 203 by the pattern matching unit 204. The degree of similarity with each command is calculated and output to the command processing unit 206. In this case, when the data completely matches the registered dictionary data, the similarity is set to 100.

For example, when voice data such as “black and white” is input, the degree of coincidence with “black and white” of the audio dictionary data in the dictionary data storage unit 203 is 95, the degree of coincidence with “copy” is 5, and “1”. The information such as the degree of coincidence with “sheet” is 10 and the degree of coincidence with “start” is 5, etc.
Sent to 6.

The command processing unit 206 determines that the command having the highest matching degree is a command input by voice and sends it to the command output unit 207.

The command output unit 207 sends the command to the driver unit 102 to set the main unit.

According to the combination of the above two operation flowcharts 3 and 4, when one color copy is to be made next time, if "copy" is uttered, the "copy" mode is set and the "color" mode is set.
By pressing the "color" key while pronounced "", a "color" mode is set, and a voice command "color" is newly registered.

Thereafter, by uttering "one sheet" and "start", a utterance command of "color" is newly registered, and a color copy operation can be performed.

By inputting voice command data as described above, necessary commands can be registered when necessary, and by preventing commands that are not normally used from being registered, the dictionary data registration section 203 can be registered. An increase in the capacity of the storage medium to be used can be suppressed.

It should be noted that the sequential detailed description of each step indicated by the dotted arrow in each step in FIG. 4 at the time of executing the voice command according to the present embodiment is as shown in the figure, and a repeated description of the sequential details will be omitted.

(Second Embodiment) A second embodiment according to the present invention will be described with reference to FIGS. 1 and 2, respectively.
Processing when a command that is not registered as a voice command is input is as follows.

First, when a voice command is input in a state where the key FIG. 8 of the operation unit 101 in FIGS. 1 and 2 is not pressed, it can be determined that a new voice command is not registered.

Next, the putter matching unit 2 in FIG.
The pattern matching with the voice dictionary data registered by the command 04 is performed, and the similarity with each command is calculated and sent to the command processing unit 206.

In this case, since the command is different from the registered command, all the similarities have low values. If they completely match, and if the similarity is set to 100, it is determined that 80 or more is a command candidate and if less than 80 is an input of an unregistered command, all matching is performed and registered. There is no command candidate.

In such a case, it is determined that the voice data input by the command processing unit has not been registered, and the command output unit 207 displays “unregistered key on the liquid crystal display unit of the operation unit 101. Please press. "
Is displayed, information is sent to the driver unit 102 so as to notify the user.

After confirming that the user has not been registered, the user presses a key on the operation unit 101 to register a new command. At the same time, the command is executed and the process proceeds to the next processing.

The operation sequence flowchart at the time of voice registration according to the second embodiment corresponds to the portion indicated by the solid line arrow in FIG. FIG. 2 shows the voice input (step S501).
Of the operation unit 10 at that time.
1 is detected by the operation unit status input unit 202, and it is determined whether or not key input is performed simultaneously (step S).
502).

If the key input is not performed at the same time, the mode is the voice data execution mode (step S503). The pattern matching of the voice data is performed by the pattern matching unit 204 (step S504). If there is no similar command (step S506), the operation unit 1 notifies the user that the command has not been registered as a message.
01 is displayed (step S506), and the key input is registered as a voice command (step S507). At the same time, the command is executed (step S508).

As described above, by inputting voice command data, when a command not registered as a voice command is input by voice, it can be easily registered as a new command.

When registering a voice command according to the second embodiment, the detailed description of each step indicated by the dotted line arrow in each step of FIG. 5 is as shown in the drawing, and the repeated description of the sequential details is omitted.

(Third Embodiment) Next, a third embodiment of the present invention will be described with reference to FIGS. 1 and 2. In the case of registering a voice command, a command already registered is Further, the operation when registration is attempted will be described.

First, when a voice command is input while the key of the operation unit 101 is pressed, it is determined that a new voice command has been registered.

Next, the pattern matching unit 204 performs pattern matching with the registered voice dictionary data, calculates the similarity with each command, and sends it to the command processing unit 206.

In this case, if the command is already registered, a command having a high similarity value exists.

In such a case, the command output unit 207 displays a command having a high degree of similarity on the liquid crystal display unit (FIG. 8) of the operation unit 101, thereby prompting the user to confirm. Send information to 102.

To give a specific example, if the user tries to register again, even though the command of “1” has already been registered, the user presses the key “1” of the operation unit 101 (FIG. 8). "".

Similar to the method described in the first embodiment, when the pattern matching is performed by the voice recognition unit 107 and the similarity is calculated, the similarity of the command “one image” becomes 90, and is equal to or more than 80. Is a command candidate,
If it is determined that less than 80 is an input of an unregistered command, it is determined that the command is already registered.

In such a case, as shown in an example in FIG. 9, the display section of the liquid crystal touch panel (FIG. 8) asks, "Is the current command" one? " ”And“ YES ”,“ N
A display for selecting "O" is made, and the user is notified that the data is similar to the registered data.

By pressing the "YES" key, the user can confirm that the user has already been registered, and
The driver unit 102 receives the input command and performs subsequent processing. In the case of "NO", re-registration is performed.

The above operation sequence flowchart is as follows.
This corresponds to the portion indicated by the solid line arrow in FIG. Voice input (step S601) is performed by the voice data input unit 201, and the state of the operation unit 101 at that time is determined by the operation unit information input unit 2
02, it is determined whether or not key input is performed simultaneously (step S602).

If the key input is performed at the same time, the operation enters the voice data registration mode (step S603). The pattern matching unit 204 performs pattern matching of the voice data (step S604). If there is a similar command (step S605), the operation unit 10 notifies the user that the command has been registered as a message.
From step 1 (step S606), if there is a key input of "YES" (step S607), the command is executed (step S608).

As described above, by performing voice command data processing, it is possible to prevent double registration when an already registered command is to be further registered.

The sequential detailed description of each step shown by the dotted arrow in each step in FIG. 6 at the time of registering the voice command according to the third embodiment is as shown in the drawing, and the repeated detailed description of the sequential details will be omitted.

(Fourth Embodiment) Next, a fourth embodiment according to the present invention will be described with reference to FIGS. 1 and 2 respectively. When an unregistered command is to be registered, it is already registered. An operation when voice data similar to the command being input is input will be described.

First, when a voice command is input in a state where a key of the operation unit 101 is pressed, it is determined that a new voice command is to be registered.

Next, the pattern matching unit 204 performs pattern matching with the registered voice dictionary data, calculates the similarity with each command, and sends it to the command processing unit 206.

In this case, when there is a command having a high similarity value among the already registered commands, the command output unit 207 displays the high similarity value on the liquid crystal display unit (FIG. 8) of the operation unit 101. Driver unit 102 so as to prompt the user for confirmation by displaying the command
Send information to

As a specific example, when a new command “8 sheets” is to be registered, the user operates the operation unit 10.
While pressing the key 1 (FIG. 8), "8" is pronounced. However, depending on the way of voice, already registered "1"
It is assumed that the similarity increases with the voice command “Sheet”.

Similar to the method described in the first embodiment, when the pattern matching is performed by the voice recognition unit 107 and the similarity is calculated, the similarity of the command “one image” becomes 85, and the command is “80” or more. Is a command candidate,
If it is determined that less than 80 is the input of an unregistered command, it is determined to be the same level of similarity as the already registered command.

In this case, erroneous recognition is likely to occur when an attempt is made to subsequently input a command by voice.

In such a case, as shown in an example in FIG. 10, the display section of the liquid crystal touch panel (FIG. 8) resembles "the current command is similar to" one ". Would you like to register again? Is displayed and a display for selecting "YES" or "NO" is issued to notify the user that the data is similar to the registered data.

When the "YES" key is selected, re-registration is performed. When the "NO" key is pressed, the voice command is registered as it is and the driver unit 102
Accepts an input command and performs subsequent processing.

The flow chart of the above operation sequence can be the same as the solid line portion in FIG. 6 of the third embodiment.
The voice input (step S601) is performed by the voice data input unit 20.
Then, the status of the operation unit 101 at that time is detected by the operation unit status input unit 202, and it is determined whether or not key input is performed simultaneously (step S602).

If the key input is performed at the same time, the operation enters the voice data registration mode (step S603). The pattern matching unit 204 performs pattern matching of the voice data (step S604). If there is a similar command (step S605), the operation unit 101 indicates to the user that a similar command has been registered as a message. (Step S606) Then, the key that has been keyed is registered as a command corresponding to the voice data (Step S607), and the command is executed (Step S608).

As described above, by performing the voice command data processing, even when the voice data to be newly registered is similar to the already registered command, the registration can be performed accurately.

In the fourth embodiment, the details of the steps indicated by the dotted arrows in each sequence will not be described again in the same manner as in the third embodiment.

(Fifth Embodiment) A fifth embodiment according to the present invention will be described with reference to FIGS.
When recognizing a voice command, a plurality of similar command candidates may appear for a command input by voice, and an operation contrary to the user's will may be performed. The following processing is performed to prevent such a malfunction.

First, when a voice command is input in a state where a key of the operation unit 101 (FIGS. 2 and 8) is not pressed, it is determined that the setting is based on the voice command.

Next, the pattern matching unit 204 (FIG. 2) performs pattern matching with the registered voice dictionary data to calculate the similarity with each command.
The command is sent to the command processing unit 206.

In this case, when a voice input similar to the command to be actually set is made, there are a plurality of commands having a high similarity.

In such a case, the command output unit 207 displays a plurality of commands having a high degree of similarity on the liquid crystal display unit of the operation unit 101, thereby instructing the driver unit 102 to prompt the user to make a selection. Send.

As a specific example, the user wants to set a command of “one sheet”, and the user pronounces “one sheet”.

However, it is assumed that the similarity with the voice command "8" is increased due to a difference in pronunciation from that at the time of registration.

Similar to the method described in the first embodiment, when the pattern matching is performed by the voice recognition unit 107 and the similarity is calculated, the similarity of the command “one image” becomes 90, and Assume that the similarity of the command “8” becomes 85. If it is determined that the similarity is 80 or more as a command candidate, and if the similarity is less than 80 as an input of an unregistered command, it is determined that both are listed as command candidates.

In such a case, as shown in an example in FIG. 11, the display section of the liquid crystal touch panel displays "Which command is it now? 1:" 1 ", 2:" 8 ". And ask the user to select either command. The driver unit 102 performs the subsequent processing using the setting selected thereby as an input command.

The above operation sequence flowchart is as follows.
The operation sequence shown in FIG. (Corresponding to a part of a portion indicated by a broken arrow in FIG. 6 of the third and fourth embodiments). Voice input (step S701) is performed by the voice data input unit 201 (FIG. 2), and the status of the operation unit 101 at that time is detected by the operation unit status input unit 202, and it is determined whether or not key input is being performed simultaneously. Perform (Step S70)
2). If the key input is not performed at the same time, the mode is the voice data execution mode (step S703). Pattern matching of voice data is performed by the pattern matching unit 204 (step S704). If there are a plurality of similar commands (steps S705 and S706), a message to the user is issued to the effect that a plurality of similar commands have been registered. From the operation unit 101 (step S70).
7) Then, the command specified by the key input (step S708) is executed as an execution command (step S709).

As described above, by performing voice command data processing, accurate setting can be performed even when a plurality of voice commands having a high degree of similarity to the input voice command are registered.

[0106]

As described above, according to the present invention,
In the image processing apparatus which operates by recognizing the command by voice, when registering data in the dictionary data storage means, a command which requires voice data to be registered while actually operating is registered when necessary. By preventing the registration of commands that are not normally used, the capacity of the storage medium used for the dictionary data storage means can be reduced.

When registering voice data, if the user tries to register the same data as the already registered data in the data of the dictionary data storage means, the fact is displayed to the user, thereby making the data double-ready. Data registration can be prevented.

Further, when an unregistered command is to be set by voice when setting by voice, it is displayed to the user that the command is unregistered, and registration can be performed each time. Become.

Further, when a command input by voice when making settings by voice is similar to a plurality of commands, those commands are displayed, and the user is prompted to make a selection, so that an accurate command is displayed. Input is possible.

[Brief description of the drawings]

FIG. 1 is a block diagram illustrating a configuration of a copying machine according to each embodiment.

FIG. 2 is a block diagram showing an internal processing configuration of a speech recognition unit in FIG. 1;

FIG. 3 is a flowchart showing an operation sequence when registering a voice command according to the first embodiment;

FIG. 4 is a diagram showing an operation sequence flowchart when a voice command is executed in the first embodiment.

FIG. 5 is an operation sequence flowchart when registering a voice command according to the second embodiment.

FIG. 6 is an operation sequence flowchart when registering a voice command according to the third and fourth embodiments.

FIG. 7 is an operation sequence flowchart at the time of registering a voice command according to a fifth embodiment.

FIG. 8 is an example of a plan view of an operation unit of the copying machine.

FIG. 9 is a display example of a liquid crystal touch panel of the operation unit according to the third embodiment.

FIG. 10 is a display example of a liquid crystal touch panel of an operation unit according to a fourth embodiment.

FIG. 11 is a display example of a liquid crystal touch panel of an operation unit according to a fifth embodiment.

[Explanation of symbols]

 Reference Signs List 101 operation unit 102 driver unit 103 image input unit 104 image processing unit 105 image output unit 106 voice input unit 107 voice recognition unit 201 voice data input unit 202 operation unit information input unit 203 dictionary data storage unit 204 pattern matching unit 205 driver information input Unit 206 command processing unit 207 command output unit

──────────────────────────────────────────────────の Continued on the front page (51) Int.Cl. ⁷ Identification code FI Theme coat ゛ (Reference) H04N 1/00 G06F 3/16 320B // G06F 3/16 320 B41J 29/00 J F-term (Reference) 2C061 AP01 AP03 AP04 AP07 AQ05 AQ06 CQ02 CQ04 CQ24 CQ25 2H027 GA19 GA20 GA25 GA26 ZA07 5B021 AA05 AA19 CC05 PP04 PP06 QQ02 QQ04 5C062 AA02 AA05 AB02 AB11 AB20 AB25 AB26 AB41 AB42 AC02 AC03 BA05 AC21 AF01 GG

Claims (6)

[Claims] An operation means for inputting a command such as a copy operation or a facsimile transmission by a switch, a button, a numeric keypad, a liquid crystal touch panel or the like, and a processing device such as a CPU. Driver means for setting processing, setting the number of prints, etc., and controlling the operation of image input means, image processing means, image output means, etc .; and An image input means for converting image data into digital data and outputting the digital data to an image processing unit; a digital image input from the image input means by a hardware circuit such as an ASIC or a software processing using the CPU or the like Image processing means for processing data in accordance with image processing settings from the driver means, and outputting the data to the image output means; An image output unit that outputs image data processed by the image processing unit by a printer such as a laser printer or an inkjet printer, and this part is used for a CRT monitor, a liquid crystal monitor, or the operation unit. Display on the liquid crystal touch panel,
An image output means including an output to a network as image data and a facsimile transmission through a modem, and a voice input using a microphone or the like, and a voice input to be sent to the voice recognition means as the digital data. And voice recognition means for inputting voice data from the voice input means, registering in dictionary data, performing voice pattern matching, converting to a command, etc., and performing command recognition by voice. A voice command recognition image processing device that operates. 2. The voice recognition means, voice data input means for inputting digital voice data from the voice input means, operation part information input means for inputting command information input from the operation means, By associating the operation unit information with the voice data,
Dictionary data storage means stored in a storage medium such as a memory or a hard disk; pattern matching between data from the voice data input means and voice dictionary data from the dictionary data storage means; A driver information inputting means for inputting the dictionary information registration and normal copy operation information from the driver means, information from the pattern matching means, and information from the command information inputting means. In the case of normal copying, the command processing unit sends a command to the command output unit by determining which command the input voice command corresponds to, and the command information from the command processing unit is copied normally. In the case of, the command output means to be sent to the mode display part of the driver means or operation means The voice command recognition image processing device according to claim 1, comprising: a stage. 3. The voice command recognition image processing device,
3. The voice command recognition image processing device according to claim 1, wherein when registering data in the dictionary data storage unit, the voice data is registered while actually operating. 4. The voice command recognition image processing device,
3. When registering the voice data, if an attempt is made to register the same data as data already registered in the dictionary data storage means, the fact is notified to a user. 3. The voice command recognition image processing device according to any one of 3. 5. The voice command recognition image processing device,
The voice according to any one of claims 1 to 4, wherein, when the setting is performed by the voice, when an unregistered command is set by the voice, the user is notified of the fact that the command is not registered. Command recognition image processing device. 6. The voice command recognition image processing device,
6. The method according to claim 1, wherein, when the setting is performed by the voice, if the command input by voice is similar to a plurality of commands, the commands are notified and the user is prompted to make a selection. The voice command recognition image processing device according to any one of the above.