JPH1166006A - System and method for managing information processor - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a system and a method for managing information processor by which a manager can be authenticated by speech recognition. SOLUTION: A speech recognition board 20 generates a feature vector by extracting features from a 'nameβ inputted in voice, calculates a distance to the standard vector of each registered person in the voice data of previously registered manager. After cutting data which are less than a threshold corresponding to the contents of managing processing inputted in voice by rearranging the voice data distance in ascending order. When the inputted 'name' is included in voice data after cutting, it is authenticated that the person inputting voice is a registered manager, and the managing processing of an information processor is accepted.

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 management system, and more particularly to an information processing apparatus management system for managing information processing apparatuses by voice.

[0002]

2. Description of the Related Art FIG. 15 illustrates a control device and a control method of a computer security system disclosed in Japanese Patent Application Laid-Open No. 4-218860. This conventional method will be described.

In FIG. 15, a user terminal device 211 is shown.
Is, for example, a personal computer or terminal device or other peripheral device having a keyboard and display device, and establishes a data line 216 to the program operation processor 212 in a conventional manner.

As shown in FIG. 15, the data line 216 must be established through a private line, a private network, a public communication network, or a public switched communication network using a modem. Can be.

[0005] In response to the start of operation of the bidirectional data line by the user terminal device 211, the program operation processor 212 transmits a "login command" response prompt to the user of the terminal device 211. Then this user,
The user inputs a login command for identifying himself, and then the program operation processor 212 prompts the user to enter a “password”.

If the login command information and password information entered by the user match the login command information and password information stored in advance and held in the program operation processor 212, the program operation processor Reference numeral 212 denotes a user's telephone 213, ideally, the user terminal device 2
Voice line 21 to telephone 213 physically closest to 11
Attempt to establish 7.

[0007] The attempt to establish the voice line 217 is based on a pre-assigned telephone number for the user,
Alternatively, it is made to any of the telephone numbers input by the user of the terminal device 211 in response to an inquiry from the program operation processor 212.

After the voice line 217 is established, the program operation processor 212 generates a four-digit random number and controls the voice device 214 to repeat the four-digit random number to the user's telephone 213. Ask the user to do so.

Next, the program operation processor 21
2 matches the received voice embodying the four-digit number with the user's stored reference voice for that four-digit number. A credit recognition value indicating the closeness of the correspondence between the received voice pattern and the stored reference voice pattern is assigned to the received voice.

This credit recognition value is then compared to a predetermined threshold established to identify a valid "voice password". If the credit recognition factor is greater than the threshold, the user can access the program operation processor 212. The voice line 217 to the telephone 213 is no longer required and is cut off.

As a method for recognizing input information such as a voice signal, there is an information recognition method disclosed in Japanese Patent Application Laid-Open No. 2-91779. In this information recognition method, a distance between an input pattern such as a voice signal and a standard pattern registered in advance is determined, a threshold is determined based on the minimum distance, and the distance calculated by the threshold is narrowed down. The selection candidates for recognizing the pattern are output.

[0012]

As described above, in order to manage the information processing apparatus, such as acquiring the apparatus configuration information and changing the configuration, first, the input person is permitted to manage the information processing apparatus. Although it is necessary to prove that the user is an administrator, conventionally, the above-mentioned operations such as password input and voice authentication have to be performed, which is troublesome.

SUMMARY OF THE INVENTION The present invention has been made to solve the above problems, and an object of the present invention is to provide an information processing apparatus management system and an information processing apparatus management method capable of authenticating an administrator by voice recognition. To provide.

[0014]

In order to achieve the above object, an information processing apparatus management system according to a first aspect of the present invention provides an information processing apparatus management system which receives input information from an administrator who is authorized to manage information processing apparatuses. Based on the information processing device management process based on
In an information processing device management system for transmitting a processing result as needed, a voice recognition unit for recognizing voice input from a communication terminal capable of voice input, and a voice registered by the voice recognition unit and a management registered in advance. Voice authentication means for performing voice authentication as to whether or not the voice input person is an administrator by comparison processing with the voice data of the administrator, and voice data of the administrator to be subjected to the comparison processing by the voice authentication means are all registered voices. Voice data selection means for narrowing down from data, wherein the voice data selection means is for narrowing down the audio data to be subjected to the comparison processing in accordance with the content of management processing by voice transmitted from the communication terminal. Set the index.

An information processing apparatus management system according to a second aspect of the present invention performs management processing of the information processing apparatus based on input information from an administrator who is authorized to manage the information processing apparatus.
In an information processing device management system for transmitting a processing result as needed, a voice recognition unit for recognizing voice input from a communication terminal capable of voice input, and a voice registered by the voice recognition unit and a management registered in advance. Voice authentication means for performing voice authentication as to whether or not the voice input person is an administrator by comparison processing with the voice data of the administrator, and voice data of the administrator to be subjected to the comparison processing by the voice authentication means are all registered voices. Voice data selecting means for narrowing down from data, wherein the voice data selecting means sets an index for narrowing down the audio data to be subjected to the comparison processing based on the received information of the input information.

According to a third aspect of the present invention, in the information processing apparatus management system according to the second aspect, the voice data selecting means sets the index according to an input path of the input information as the received information. .

According to a fourth aspect of the present invention, in the information processing device management system according to the third aspect, the voice data selecting means is configured to input the narrowing level set as the index and the input information within a private line. It is set to be higher when input is made outside the private line than when.

According to a fifth aspect of the present invention, in the information processing apparatus management system according to the second aspect, the voice data selecting means is configured to determine the index based on the number of times of input information from each of the communication terminals as the received information. Is set.

According to a sixth aspect of the present invention, in the information processing apparatus management system according to the second aspect, the audio data selecting means sets the index according to a connection mode of the communication terminal as the received information. is there.

According to a seventh aspect of the present invention, in the information processing device management system according to the first or second aspect, the voice data selecting means is common to the voice recognized by the voice recognition means of the manager's voice data. The threshold value for discarding audio data that does not have a common feature when arranged in the order of having the feature is used as the index.

An information processing apparatus management method according to an eighth aspect of the present invention performs management processing of the information processing apparatus based on input information from an administrator who is authorized to manage the information processing apparatus. A voice recognition step of recognizing processing content by voice input from a communication terminal capable of voice input,
An index setting step of setting an index for narrowing down audio data according to the content of management processing by voice sent from the communication terminal; and a voice narrowed down based on the index among voice data of a pre-registered manager. Performing a voice authentication step of performing voice authentication as to whether or not the voice utterer is an administrator by comparing data and voice input from the communication terminal, and executing a management process in accordance with the result of the voice authentication. This is to decide whether or not to allow.

According to a ninth aspect of the present invention, there is provided an information processing apparatus management method for performing management processing of an information processing apparatus based on input information from an administrator who is authorized to manage the information processing apparatus. A voice recognition step of recognizing processing content by voice input from a communication terminal capable of voice input,
An index setting step of setting an index for narrowing down audio data based on the reception information of the input information; and, among voice data of a pre-registered manager, voice data narrowed down based on the index and input from the communication terminal. A voice authentication step of performing voice authentication to determine whether or not the voice input person is an administrator by comparing the voice with the received voice, and determining whether to execute a management process according to the voice authentication result. .

[0023]

Preferred embodiments of the present invention will be described below with reference to the drawings.

Embodiment 1 FIG. 1 is a block diagram of the entire system to which the first embodiment of the information processing apparatus management system according to the present invention is applied. The information processing apparatus 1 shown in FIG. 1 connects an extension telephone line 2 and a LAN 3. A telephone 4 and a PBX 5 are connected to the telephone line 2. FA is connected to the PBX 5 via a telephone line (external line) 6.
The X device 7 is connected. On the other hand, a management terminal 8 for managing the information processing apparatus 1 and a router 9 connectable to the Internet are connected to the LAN 3.

FIG. 2 is a block diagram of the information processing apparatus 1 according to the present embodiment. Inside the information processing device 1, a system bus 11 and an I / O bus 12 are provided. The system bus 11 includes two CPU boards 13,
14, a memory board 15 and a bus bridge board 16 are connected to each other, and perform data transfer via the system bus 11. In addition to the bus bridge board 16 for connecting to the system bus 11, two I / O buses 12 are connected to two SCSI control boards 17, 18, a serial communication board 19 and a voice recognition board 20. Each SC
The SI control boards 17 and 18 are SCSI buses 21 and 22
The disk units 23 and 24 are connected via the. In this embodiment, the voice recognition board 20 operates as a voice recognition means for recognizing voice input from a communication terminal capable of inputting voice, and operates as a voice recognition means to control the voice recognized by the voice recognition means in advance. The voice data of the administrator to be subjected to the comparison processing performed as voice authentication means as voice authentication means for performing voice authentication as to whether or not the voice input person is an administrator by comparison processing with the voice data of the user. It is provided as audio data selecting means for narrowing down from registered audio data.

Further, the information processing apparatus 1 is provided with a system management bus 25 provided for system management, and includes CPU boards 13 and 14, a memory board 15, a bus bridge board 16, and SCSI control boards 17 and 18. ,
A system management board 26 mounted for controlling and managing the information processing device 1 is connected via the disk units 23 and 24 and the system management bus 25. In FIG. 2, the system management bus 25 is represented by a broken line for convenience. The system management board 26 is connected to the I / O bus 12 via the serial communication board 19,
It is also connected to a telephone line. Further, the information processing apparatus 1 has a configuration not directly related to the execution of the processing function, but includes a power supply unit 27 for supplying power to the main body,
A fan unit 28 for cooling the main body is mounted. Note that the information processing device itself may have the same configuration as the conventional one.

In this embodiment, the configuration and function of the system management board 26 mounted on the information processing apparatus 1 are characterized.
6 is a block diagram showing the configuration of FIG.

In FIG. 3, the system management board 26
Is a processor 31 for executing functions of the board 26 by executing an application stored in the ROM 32 or the RAM 33, and a serial communication circuit 34 connected to the serial communication board 19 and communicating with the information processing apparatus 1. And a telephone line circuit 35 connected to the telephone line and communicating via the telephone line. Further, the system management board 26 includes a voice detection circuit 36 for detecting telephone tones and voices transmitted from a telephone line, a voice control circuit 37 for responding to the received telephone tones and input information, and a message and processing to be transmitted. A voice data memory 38 for temporarily storing data such as results, and a voice synthesis circuit 39 for converting the data into voice
And a speech encoding circuit 40 for encoding speech received via the telephone line circuit 35 and the speech detection circuit 36 and storing the encoded speech in the speech data memory 38.

The system management board 26 according to the present embodiment includes a processor 31, a ROM 32, a RAM 3
3. The contents of the input information sent from outside can be recognized by the telephone line circuit 35, the voice detection circuit 36, and the voice control circuit 37. Further, a processor 31, a ROM 32, a RAM 3 constituting
3. A predetermined process is executed by the serial communication circuit 34 based on the input information. Further, a processor 31, a ROM 32, a RAM 33, and a telephone line circuit 3 which constitute a reporting unit
5. The voice detection circuit 36, voice data memory 38, voice synthesis circuit 39, and voice encoding circuit 40 allow the processing result of the information processing device control management means to be transmitted to an external device such as a telephone or a facsimile device specified based on the input information. Send to communication terminal. System management board 26 in the present embodiment
Has a voice conversion function of converting the processing result (data) into voice, so that the processing result can be reported to the telephone 4 by voice.

In this embodiment, the system management bus 25 is IEEEStd. 1149.1 (JTA
An example using a G) -compliant serial bus is shown. Therefore, according to the present embodiment, IEEE Std. 1149.1
CPU boards 13, 14, memory board 15, bus bridge board 16, SCSI control boards 17, 18, serial communication board 19, disk unit 23 on a (JTAG) compliant serial bus (hereinafter referred to as "JTAG bus"). , 24 and the system management board 26 are connected in a daisy chain, each having a different address used in the protocol of the JTAG bus. The same applies to each of the embodiments described below.

Next, the operation in the present embodiment will be described with reference to the flowchart shown in FIG. 4. First, an instruction for processing of acquiring the state of the information processing apparatus 1 from the telephone 4 connected to the telephone line 2 will be given. Will be described by taking as an example a case in which the sound is made by voice. The state acquisition processing refers to the state of each component shown in FIG. 2, for example, the operation state of the CPU and the memory, the operation state of the fan, and the monitoring state of a temperature sensor and a voltage sensor (not shown). This is the process to acquire.

First, when a call is made to the information processing apparatus 1 from the telephone 4 connected to the telephone line 2, the system management board 26 connected to the telephone line 2 receives the telephone tone (access request) (step). 101). When the voice detection circuit 36 detects the telephone tone via the telephone line circuit 35, the voice control circuit 37 responds to the telephone tone. Here, the processor 31 includes a ROM 32
Alternatively, by executing a predetermined application stored in the RAM 33, a predetermined “response message” (data) is voice-synthesized by the voice synthesis circuit 39. Then, the processor 31 transmits the "response message" (voice) obtained by the voice synthesis to the telephone 4 connected to the telephone line 2 via the voice detection circuit 36 and the telephone line circuit 35 (step 102). The “response message” (voice) is, for example, “Please describe the processing content to access the information processing device.”

In response to the voice response message, the voice input person inputs "status" (voice) from the telephone 4 connected to the telephone line 2 in order to obtain the status of the information processing apparatus 1. The voice encoding circuit 40 determines the “state” (voice)
Is received via the telephone line circuit 35 and the voice detection circuit 36 (step 103), the voice input information is encoded and stored as "state" (data) in the voice data memory 3.
8 is stored. Processor 3 of system management board 26
1 stores this “state” (data) in the audio data memory 3
8 to the voice recognition board 20 via the serial communication circuit 34 and the serial communication board 19 of the information processing apparatus 1. The voice recognition board 20 receives this "state" (data), extracts a feature and generates a feature vector, and then calculates a distance with respect to a standard vector of each command registered in the command data. As a result, the "state"
Since (Data) has the shortest distance, the input voice is recognized as "state" and is converted to "state" (command) (step 104). The command data is data in which a command group prepared for the purpose of managing the information processing device 1 is registered in advance, and the command data is stored in the voice recognition board 20 or the disk unit 23.
24 are registered. The voice recognition board 20 again
The “state” (command) is transmitted to the audio data memory 38 via the serial communication board 19 and the serial communication circuit 34 of the system management board 26. The processor 31 of the system management board 19 processes the "state" (command), synthesizes the "response message" (data) by voice by the voice synthesis circuit 39, and converts the "response message" (voice) to the voice detection circuit 36. Is transmitted to the telephone 4 connected to the telephone line 2 via the telephone line circuit 35 (step 105). This "answer message" (voice)
Is, for example, "Please say your registered" name "."

In response to the voice response message, the voice input person inputs a "name" (voice) from the telephone 4 for authentication. When the voice encoding circuit 40 receives the voice input information "name" (voice) via the telephone line circuit 35 and the voice detection circuit 36 (step 10).
6) Encode and store in the audio data memory 38 as "name" (data). The processor 31 of the system management board 26 transmits the “name” (data) and the registrant data from the voice data memory 38 via the serial communication circuit 34 and the serial communication board 19 of the information processing device 1.
The data is transmitted to the voice recognition board 20. The registrant data is voice data of a manager who is permitted to manage the information processing apparatus 1 registered in advance. In the present embodiment,
Although the registrant data is stored in the voice data memory 38, it may be stored as a file in another storage means such as the RAM 33 or an external storage device. When receiving the "name" (data), the voice recognition board 20 performs the following processing to authenticate the administrator (step 10).
7). FIG. 5 shows a flowchart of the administrator authentication processing.

First, the voice recognition board 20 has a "name"
After extracting features from (data) and generating feature vectors, distance calculation is performed on the standard vector of each registrant in the registrant data, and sorting is performed in ascending order of distance (step 121). For the registrant data sorted in ascending order, a threshold based on “processing content: acquisition of status” is set (step 122), and data that does not satisfy the threshold is cut off (step 123). Note that the smaller the number, the more similar the distance calculation result means the similarity. Therefore, the fact that the value is less than the threshold means that the number is larger than the threshold. For example, if the processing content is state acquisition, the threshold for narrowing down in speech recognition is set to 3
If the calculation result does not reach the threshold value 3,
The registrant data of 5, ... is truncated.
In this embodiment, the threshold value is set as an index for narrowing the registrant data to be compared with the input voice data "name" in this manner. In the case of the instruction of "acquisition of information" dealt with here, unlike the case of "configuration change", a security breach because the management process to be executed does not affect the system configuration by merely collecting information It is considered that it is advantageous to give priority to improving the recognition efficiency by increasing the threshold value rather than the recognition accuracy. That is, for example, by raising the threshold from 4 to 3 (the number becomes smaller), the number of registrant data to be truncated out of the total number of registrant data increases, and the number of registrant data to be compared becomes smaller and narrowed down. The time required for the comparison process can be shortened and the efficiency can be improved. On the other hand, for example, by lowering the threshold from 4 to 5 (the number increases), the number of registrant data to be truncated out of the total number of registrant data relatively decreases, so that the number of registrant data to be compared is increased. However, the recognition accuracy can be improved accordingly. Therefore, it can be said that the probability of erroneously recognizing a non-administrator as a manager is low. In the case of acquiring information, the effect of the security breach is relatively small, so that recognition efficiency is prioritized over recognition accuracy, and the threshold for narrowing down is set to 3.

If the input "name" (data) is included in the truncated registrant data of the voice data less than the threshold, the voice creator is authenticated as a registered manager. (Steps 124 and 125). On the other hand, if it is not included in the registrant data after the truncation, it is determined that the voice input person is not a registered manager (step 12).
4,126).

In FIG. 4, when the voice recognition person is determined to be the registered manager in the manager authentication process (step 108), the serial communication board 19 and the system management board 26 Via the serial communication circuit 34 to the audio data memory 38. The processor 31 of the system management board 26
The status of each component of the information processing device 1 is monitored via the JTAG bus 25, and the status of each component is put together (step 109). The processor 31 of the system management board 26 outputs “No failure” (data) if there is no failure in each component of the information processing apparatus 1, “Failure location and contents” (data) if there is at least one failure, It is created on the voice data memory 38 and converted into "no failure" (voice) or "failure location, contents" (voice) by the voice synthesis circuit 39 (step 110), and passed through the voice detection circuit 36 and the telephone line circuit 35. Then, the data is transmitted to the telephone 4 connected to the telephone line (step 111). "Fault location, content" (voice)
Is, for example, “AC / DC unit # 1, output voltage drop”. If it is determined that the voice input person is not a registered administrator, the established line is forcibly disconnected. As described above, in the present embodiment, whether to execute the management process is determined according to the result of the voice authentication.

Next, a case will be described as an example in which the telephone 4 connected to the telephone line 2 gives a voice instruction to change the configuration of the information processing apparatus 1. The configuration change processing refers to processing such as degeneration of a processor, a memory, and a disk, or addition of a memory. here,
An example will be described in which the processor is degenerated. Note that the processing flow here is the same as above, except that the content of the management processing has changed from acquiring the status to changing the configuration.

That is, in this processing, after detecting a telephone tone, a "response message" (voice) "Please describe the processing contents in order to access the information processing apparatus" is transmitted to the telephone 4. The processing up to is the same as the above (steps 101 and 102). In addition, the voice input person inputs “change” (voice) from the telephone 4 connected to the telephone line 2 in order to change the configuration of the information processing apparatus 1 in response to the voice response message. The process up to the transmission of the "response message" (voice) saying "name" to the telephone 4 is the same as the above, except that the voice to be handled is "change" instead of "state" ( Steps 103 to 105). Then, the voice entrant responds to the voice response message,
A "name" (voice) is input from the telephone 4 for authentication. The voice input information "name" (voice) is converted into a "name" (data) and transmitted to the voice recognition board 20 together with the registrant data. The processing up to this point is the same as above (step 106). Here, the administrator authentication process performed by the voice recognition board 20 receiving the “name” (data) will be described.

In FIG. 5, the voice recognition board 20
After extracting features from “name” (data) and generating a feature vector, distance calculation is performed on the standard vector of each registrant in the registrant data, and sorting is performed in ascending order of distance (step 121). For the registrant data sorted in ascending order, a threshold based on “processing content: change of configuration” is set (step 122), and those less than the threshold are cut off (step 123). For example, when the processing content is a configuration change, the threshold value for narrowing down the voice recognition is set to 5. That is, in this case, the effect of the security breach is greater than in the case of acquiring the state described above, and it is advisable to give priority to improving the recognition accuracy by lowering the threshold value rather than the recognition efficiency. Therefore, in the case of changing the configuration, the threshold for narrowing down is set to 5, which is larger than 3 in the case of acquiring information.

As described above, by making the threshold value relatively large, the number of registrant data to be truncated decreases, and the number of data to be compared with the input “name” (data) increases. Although the recognition efficiency is relatively inferior, the recognition accuracy can be improved accordingly.

If the input "name" (data) is included in the truncated registrant data of the voice data less than the threshold, the voice creator is authenticated as a registered manager. (Steps 124 and 125). On the other hand, if it is not included in the registrant data after the truncation, it is determined that the voice input person is not a registered manager (step 12).
4,126).

In FIG. 4, when the voice recognition person is judged to be the registered manager in the manager authentication process (step 108), the serial communication board 19 and the system management board 26 are again activated. Via the serial communication circuit 34 to the audio data memory 38. Then, in the case of a configuration change, the following processing is executed (step 109).

That is, the processor 31 of the system management board 26 executes an application corresponding to the “change” (character code), and causes the voice synthesis circuit 39 to perform voice synthesis of the “configuration change menu” (character code). Then, the processor 31 transmits the “configuration change menu” (voice) obtained by the voice synthesis to the telephone 4 via the voice detection circuit 36 and the telephone line circuit 35. This “configuration change menu” (voice) is, for example, “Please say the components to be changed and the details of the change to change the configuration of the information processing device.”

The administrator can change the configuration of the information processing apparatus 1,
In this example, in order to degenerate the CPU board 13, a voice input “processor, degeneration” (voice) is input from the telephone 4. This “processor, degenerate” (voice) is encoded by the voice encoding circuit 40 via the telephone line circuit 35 and the voice detection circuit 36 and stored in the voice data memory 38 as “processor, degenerate” (data). Is done. The processor 31 of the system management board 26 transmits this “processor, degenerate” (data) from the voice data memory 38 to the voice recognition board 20 via the serial communication circuit 34 and the serial communication board 19 of the information processing device 1. I do.
The voice recognition board 20 receives the “processor, degenerate” (data), extracts a feature, generates a feature vector, calculates a distance to a standard vector of each command in the command data, and calculates a distance that is the smallest. Convert to "processor, degenerate" (command). The voice recognition board 20
Again, the serial communication board 19 and the system management board 2
6 to the audio data memory 38 via the serial communication circuit 34.

The processor 31 executes a predetermined application on the basis of the “processor, degenerate” (character code) and sends the application to the CPU board 13 via the JTAG bus 25.
For example, after disconnecting a failed processor from among them, the information processing apparatus 1 is restarted and its operation is confirmed. In this way, the administrator can easily change the configuration of the information processing apparatus 1 by giving a voice instruction from the telephone 4.

As described above, according to the present embodiment, by making a call to the information processing device 1 from the telephone 4 connected to the telephone line 2, the information processing device 1 performs authentication by voice of the telephone. When it is recognized by this authentication function that the administrator is a registered voice, the status of the information processing apparatus 1 can be acquired by voice of the telephone or the configuration can be changed. By setting the threshold for narrowing down the authentication function by voice to be higher (larger number) in the case of changing the configuration than in the case of acquiring the state, when changing the configuration that requires a higher security level Can be relatively improved. On the other hand, it is possible to relatively increase the recognition efficiency when acquiring a state that does not require a higher security level. In the present embodiment, the index for narrowing down the registrant data, that is, the threshold, can be changed, so that the recognition accuracy or the recognition efficiency can be improved according to the content of the processing.

In the present embodiment, two management processes of acquiring a state and changing a configuration are taken as examples of management processes, and a threshold value for narrowing down speech recognition can be changed for each management process. However, threshold values are set for other management processes prepared in consideration of recognition accuracy and recognition efficiency.

Each of the thresholds may be set in advance in the speech recognition board 20 in the form of a table corresponding to each management process. Of course, the setting may be made in a storage unit such as a memory provided in another unit or an external storage device such as a disk unit.

Further, in the present embodiment, as shown in FIG. 4, a response message for inquiring about the processing contents and the name is created and separately input by voice. Is also good.

Further, the system management bus is IEEEES.
td. Although a serial bus conforming to 1149.1 (JTAG) is used, it can be realized by a serial bus such as I2C or a parallel bus. The same applies to the following embodiments.

Embodiment 2 In the first embodiment, an index, that is, a threshold value, for narrowing down audio data to be compared can be set according to the content of the management processing of the information processing apparatus 1. Each of the embodiments after this embodiment is characterized in that a threshold value is set based on reception information of input information by voice. The present embodiment is characterized in that a threshold value is set according to an input path of input information as reception information.

FIG. 6 is a block diagram of the whole system to which the second embodiment of the information processing apparatus management system according to the present invention is applied. In the present embodiment, the telephone 41 is connected to the telephone line (external line) 6 having the configuration shown in FIG. The voice recognition board 20 in the present embodiment operates as the voice recognition means and the voice data selection means, and further operates as the voice recognition means to perform a comparison process between the voice recognized by the voice data of the administrator registered in advance. It is provided as a voice authentication unit that performs voice authentication as to whether the voice input person is an administrator. Note that the block configuration of the information processing apparatus and the block configuration of the system management board according to the present embodiment are the same as those of the first embodiment as shown in FIGS.

In the present embodiment, the voice input from the telephone 41 connected to the telephone line (external line) 6 and the voice from the telephone 4 connected to the telephone line (extension) 2 A description will be given of a case where the input is made.

First, based on the flow chart shown in FIG. 7, an example will be described in which the telephone 41 connected to the telephone line (outside line) 6 gives a voice instruction to acquire the state of the information processing apparatus 1. I do.

When the telephone 41 connected to the telephone line 6 makes a call to the information processing apparatus 1, the system management board 26 connected to the telephone line 6 receives the telephone tone (access request) (step 201). . When the voice detection circuit 36 detects the telephone tone via the telephone line circuit 35, the voice control circuit 37 responds to the telephone tone. Here, the processor 31 executes a predetermined application stored in the ROM 32 or the RAM 33 to generate a predetermined “response message”.
(Data) is subjected to speech synthesis by the speech synthesis circuit 39. Then, the processor 31 transmits the "response message" (voice) obtained by the voice synthesis to the telephone 41 connected to the telephone line 6 via the voice detection circuit 36 and the telephone line circuit 35 (step 202). The “response message” (voice) is, for example, “Please say your registered“ name ”.”

The voice input person inputs a "name" (voice) from the telephone 4 for authentication in response to the voice response message. The voice encoding circuit 40 receives the voice input information "name" (voice) via the telephone line circuit 35 and the voice detection circuit 36 (step 20).
3) Encode and store in the audio data memory 38 as "name" (data). The processor 31 of the system management board 26 transmits the “name” (data) and the registrant data from the voice data memory 38 via the serial communication circuit 34 and the serial communication board 19 of the information processing device 1.
The data is transmitted to the voice recognition board 20. Voice recognition board 20
Receives the "name" (data), and performs the following processing to authenticate the administrator (step 204).
FIG. 8 shows a flowchart of this administrator authentication processing.

First, the voice recognition board 20 has a “name”
After extracting features from (data) and generating feature vectors, distance calculation is performed on the standard vector of each registrant in the registrant data, and rearrangement is performed in ascending order of distance (step 221). For the registrant data sorted in ascending order, a threshold based on “access route: telephone line (external line)” is set (step 222), and those less than the threshold are discarded (step 223). For example, when the access route (input route) is from the telephone line (external line) 6, the threshold for narrowing down the voice recognition is set to 5. The reason for setting the threshold relatively high is that if the access route is a telephone line (external line) 6, access from outside the office is assumed, and the possibility that the voice input person is a person other than the administrator is compared with the inside of the office. It will be higher. Therefore, the security level should be relatively high and the authentication accuracy should be high.

If the input “name” (data) is included in the truncated registrant data of the voice data less than the threshold value, the voice creator is authenticated as a registered manager. (Steps 224, 225). On the other hand, if it is not included in the registrant data after the truncation, it is determined that the voice input person is not a registered manager (step 22).
4,226).

As described above, in the present embodiment, the input path from where the input information is issued is used as the information obtained when the received information, that is, the input information by voice is received. Whether the called telephone is connected to an outside line (outside the private line) or an internal line (inside the private line) can be known from the number notification at the time of line establishment.

In FIG. 7, when the voice recognition board 20 determines that the voice input person is not a registered manager in the manager authentication process, it disconnects the established line (step 205). On the other hand, when the voice recognition board 20 determines that the voice input person is the registered manager in the manager authentication process (step 205), the serial communication board 19 and the serial communication circuit 34 of the system management board 26 are again operated. Is transmitted to the audio data memory 38 via Here, the processor 31 is provided with the ROM 32 or the R
By executing a predetermined application stored in the AM 33, a predetermined “response message” (data) is voice-synthesized by the voice synthesis circuit 39. Then, the processor 31 transmits the "response message" (voice) obtained by the voice synthesis to the telephone 4 connected to the telephone line 2 via the voice detection circuit 36 and the telephone line circuit 35 (step 206). This "response message"
(Speech) is, for example, "Please describe the processing content to access the information processing device." In the present embodiment, whether to execute the management process is determined in accordance with the result of the voice authentication as described above.

In response to the voice response message, the voice input person inputs “status” (voice) from the telephone 41 connected to the telephone line 6 in order to obtain the status of the information processing apparatus 1. The voice encoding circuit 40 receives the voice input information of "state" (voice) via the telephone line circuit 35 and the voice detection circuit 36 (step 20).
7) The data is encoded and stored in the audio data memory 38 as "state" (data). The processor 31 of the system management board 26 transmits the “status” (data) from the audio data memory 38 to the serial communication circuit 34 and the information processing device 1.
Via the serial communication board 19 to the voice recognition board 20. The voice recognition board 20 reads the “state”
After receiving (data), extracting a feature and generating a feature vector, distance calculation is performed on a standard vector of each command registered in the command data. As a result, since the distance of “state” (data) is the shortest, the input voice is recognized as “state” and is recognized as “state”.
(Command) (step 208). The voice recognition board 20 transmits the “state” (command) again to the voice data memory 38 via the serial communication board 19 and the serial communication circuit 34 of the system management board 26. The processor 31 of the system management board 26 processes the “status” (command), monitors the status of each component of the information processing apparatus 1 via the JTAG bus 25, and puts together the status of each component ( Step 209).
If there is no failure in each component of the information processing apparatus 1, the processor 31 of the system management board 26 determines "no failure".
(Data), if there is at least one fault, “failure location, contents” (data) is created on the voice data memory 38;
The voice synthesizing circuit 39 converts the data into "no failure" (voice) or "failure location, contents" (voice) (step 21).
0), and transmits to the telephone 41 connected to the telephone line via the voice detection circuit 36 and the telephone line circuit 35 (step 211). “Fault location, content” (voice) is, for example,
“AC / DC unit # 1, output voltage drop”.

Next, a case will be described as an example in which the telephone 4 connected to the telephone line (extension) 2 gives a voice instruction for processing of acquiring the state of the information processing apparatus 1. This process is the same as the information acquisition process from the telephone 41 connected to the telephone line (extension) 6.

That is, when a telephone tone from the telephone 4 is detected, a response message (voice) "Please say your registered" name "is transmitted, and a voice entrant responding to the voice response message. Is received (steps 201 to 20)
3) The next administrator authentication process is performed (step 204).

In FIG. 8, the speech recognition board 20
After extracting features from “name” (data) and generating a feature vector, distance calculation is performed on the standard vector of each registrant in the registrant data, and rearrangement is performed in ascending order of distance (step 221). For the registrant data sorted in ascending order, a threshold based on "access route: telephone line (extension)" is set (step 222), and those less than the threshold are discarded (step 223). For example,
When the access route (input route) is from the telephone line (extension) 2, the threshold for narrowing down the voice recognition is set to 3. The reason why the threshold value is set lower than that from the telephone line (external line) 6 is that if the access route is the telephone line (extension line) 2, it is assumed that access will be made from within the office, so that the voice input person is other than the administrator. It is considered that the likelihood of being a person is lower than that from the outside. Therefore, it is better to lower the security level than in the case of the telephone line (outside line) 6 and give priority to recognition efficiency over authentication accuracy.

If the input "name" (data) is included in the truncated registrant data of the voice data less than the threshold, the voice creator is authenticated as a registered manager. (Steps 224, 225). On the other hand, if it is not included in the registrant data after the truncation, it is determined that the voice input person is not a registered manager (step 22).
4,226).

The processing after step 205 in FIG. 7 is the same as the processing from the telephone 41 connected to the telephone line (external line) 6, and the description is omitted.

As described above, according to the present embodiment, by making a call from telephones 4 and 41 to information processing apparatus 1, information processing apparatus 1 performs authentication by telephone voice, and this authentication function performs voice authentication. Is recognized as a registered administrator, the status of the information processing apparatus 1 can be acquired by telephone sound. By setting a threshold value for narrowing down the voice authentication function in the case of acquiring the status in the telephone line (external line) 6 higher than in the case of acquiring the state in the telephone line (extension line) 2, In the case of acquiring the status on the line (outside line) 6, the recognition accuracy can be relatively increased. On the other hand, in the case of acquiring the status on the telephone line (extension) 2, the recognition efficiency can be relatively increased.

In the present embodiment, the state acquisition process has been described as an example, but it goes without saying that the present invention can be applied to other management processes such as a configuration change process.

In this embodiment, a telephone line (external line) and a telephone line (extension) are taken as examples of access routes.
Thresholds for narrowing down in voice recognition can be set for each access route, but for other access routes, for example, access from inside and outside the firewall when connecting to the Internet, thresholds are individually set,
It is also possible to increase recognition accuracy or recognition efficiency.

Embodiment 3 In the second embodiment, the threshold is set as the reception information of the input information according to the input path such as the outside line or the extension of the telephone line. The present embodiment is characterized in that the number of times that input information by voice is input from each communication terminal as reception information is counted, and an index is set according to the number of inputs.

FIG. 9 is a block diagram of an entire information processing apparatus management system according to the third embodiment of the present invention. The present embodiment has a configuration in which a telephone 42 is connected to the telephone line (extension) 2 having the configuration shown in FIG.
It has the same function as. Note that the block configuration of the information processing apparatus and the block configuration of the system management board according to the present embodiment are the same as those of the first embodiment as shown in FIGS.

In the present embodiment, when the voice is input from the telephone 4 whose input frequency is counted to some extent because it is used for the maintenance of the normal information processing apparatus 1, it is used for the maintenance of the normal information processing apparatus 1. There will be described a case where a voice is input from the telephone 42, where the counted number of times of input is extremely small.

First, a case where a voice is input from a telephone 42 not used for maintenance of the information processing apparatus 1 will be described as an example. The operation in the present embodiment is the same as that in the second embodiment.
Since the processing is basically the same as that of the first embodiment, the common processing will be briefly described with reference to the flowcharts shown in FIGS.

When the telephone 42 connected to the telephone line 2 makes a call to the information processing apparatus 1, the system management board 26 connected to the telephone line 2 receives the telephone tone (access request) (step 201). Then, a response message (voice) saying "Please say the registered" name "" synthesized by the voice synthesis circuit 39 is transmitted (step 202). When the "name" (voice) from the voice input person corresponding to the voice response message is received (step 203), the next administrator authentication process is performed (step 204).

In FIG. 8, the voice recognition board 20
After extracting features from “name” (data) and generating a feature vector, distance calculation is performed on the standard vector of each registrant in the registrant data, and rearrangement is performed in ascending order of distance (step 221). For the registrant data sorted in ascending order, a threshold based on the total access count (input count) of the input destination telephone number is set (step 222).
Those that do not satisfy the threshold value are discarded (step 22).
3). For example, when the total number of accesses from the telephone 42 to which a certain telephone number is assigned is less than 10, the threshold for narrowing down the voice recognition is set to 5. If the total number of accesses is less than 10, it is better to determine that the input is a voice input from the telephone 42 that is not normally used for maintenance, to set the security level relatively high, and to increase the authentication accuracy. . It is necessary to provide a means for recording the total number of accesses received from each telephone number.
It can be realized by a storage means such as 0 or another prepared memory or an external storage device such as a disk unit.

If the input “name” (data) is included in the truncated registrant data of the voice data less than the threshold, the voice creator is authenticated as a registered manager. (Steps 224, 225). On the other hand, if it is not included in the registrant data after the truncation, it is determined that the voice input person is not a registered manager (step 22).
4,226).

The processing after step 205 in FIG. 7 is the same as that in the second embodiment, and a description thereof will be omitted.

Next, a case where a voice is inputted from the telephone 4 used for maintenance of the information processing apparatus 1 will be described as an example.

When a call is made from the telephone 4 connected to the telephone line 2 to the information processing apparatus 1, the system management board 26 connected to the telephone line 2 receives the telephone tone (access request) (step 201). The "registered" name "" which is speech-synthesized by the speech synthesis circuit 39.
Please say Is transmitted (step 202). When the "name" (voice) from the voice input person corresponding to the voice response message is received (step 203), the next administrator authentication process is performed (step 204).

In FIG. 8, the voice recognition board 20
After extracting features from “name” (data) and generating a feature vector, distance calculation is performed on the standard vector of each registrant in the registrant data, and rearrangement is performed in ascending order of distance (step 221). For the registrant data sorted in ascending order, a threshold based on the total access count (input count) of the input destination telephone number is set (step 222).
Those that do not satisfy the threshold value are discarded (step 22).
3). For example, if the total number of accesses from the telephone 4 to which a certain telephone number is assigned is 10 or more,
The threshold value for narrowing down the voice recognition is set to 3. When the total number of accesses is 10 or more, it is determined that the input is a voice input from the telephone 4 used for normal maintenance, and the security level is relatively lower than when the total number of accesses is less than 10 times. This is because it is better to increase the authentication efficiency.

If the input "name" (data) is included in the truncated registrant data of the voice data less than the threshold, the voice creator is authenticated as a registered manager. (Steps 224, 225). On the other hand, if it is not included in the registrant data after the truncation, it is determined that the voice input person is not a registered manager (step 22).
4,226).

The processing after step 205 in FIG. 7 is the same as that in the second embodiment, and a description thereof will be omitted.

As described above, according to the present embodiment, by making a call from telephones 4 and 42 to information processing apparatus 1, information processing apparatus 1 performs authentication by telephone voice, and this authentication function performs voice authentication. Is recognized as a registered administrator, the status of the information processing apparatus 1 can be acquired by telephone sound. The threshold for narrowing down the authentication function by voice is changed in accordance with the total number of accesses from the telephone, noting that the telephone used by the normal administrator will be specified in this embodiment. I did it. Thus, in the case of voice input from a frequently accessed telephone, recognition efficiency is more important than recognition accuracy, whereas when the number of accesses is small, recognition accuracy is more important than recognition efficiency.

In the present embodiment, the total number of accesses for setting the threshold value is based on ten times. However, the number of accesses is not limited to this number. It is possible to apply such as step by step. Also, the point at which accumulation of the number of accesses is started may be determined as appropriate.

In the present embodiment, the state acquisition processing has been described as an example. However, it goes without saying that the present invention can be applied to other management processing such as a configuration change processing.

Embodiment 4 FIG. 10 is a block diagram of the entire system to which the third embodiment of the information processing apparatus management system according to the present invention is applied. In the present embodiment, in addition to the configuration shown in FIG. 1, a management terminal 45 for managing the information processing device 1 is directly connected to the information processing device 1, and an Internet telephone that can be connected to the Internet via the router 9. 46 is connected to the LAN 3. In FIG. 10, an Internet telephone 46 is connected to the same LAN 3 as the information processing apparatus 1 for convenience.
The Internet telephone 46 accesses the information processing device 1 via the Internet. Further, the voice recognition board 20 has the same function as the second embodiment.

FIG. 11 is a block diagram of the information processing apparatus 1 according to the present embodiment. In the present embodiment, in addition to the configuration shown in FIG. 2, a network board 47 for connecting to the LAN 3 is connected to the I / O bus 12.

The block configuration of the system management board 26 in the present embodiment is the same as that of the first embodiment as shown in FIG.

FIG. 12 is a block diagram of an Internet telephone 46 according to the present embodiment. The Internet telephone 46 is an Internet telephone means that can access the Internet and can input and output both text and voice. The Internet telephone 46 has a configuration similar to that of a general computer, and includes a CPU 49 and a memory 50 connected to a system bus 48, a SCSI controller 52 connected to an I / O bus 51, a LAN board 53, and a sound circuit. 54, an I / O bridge 55 connecting the system bus 48 and the I / O bus 51, and a SCSI controller 52
And a disk unit 56 that is connected and managed by the server. The Internet telephone 46 is connected to the Internet via a LAN board 53. In addition, the sound circuit 54 has a voice conversion function of converting received data into voice, and can input voice of an administrator or the like from the connected headphone set 57 and send the voice from the information processing apparatus 1. It is possible to hear the coming processing result and the like by voice. Not only voices from the Internet telephone 46 but also character data (not shown) can be input from a keyboard.
3 to the information processing apparatus 1.

FIG. 13 is a block diagram of the management terminal 45 in the present embodiment. The management terminal 45 has the same configuration as a general computer. That is,
CPU 61 and memory 6 connected to system bus 60
2 and a SCSI controller 6 connected to the I / O bus 63
4, a serial communication circuit 65, a sound circuit 66, an I / O bridge 67 connecting the system bus 60 and the I / O bus 63, and a disk unit 68 connected and managed by the SCSI controller 64. The management terminal 45
The serial communication circuit 65 is connected to the serial communication board 19 of the information processing apparatus 1. In addition, sound circuit 6
Reference numeral 6 has a voice conversion function of converting received data into voice, and can input voice of an administrator or the like from the connected headphone set 69 and process results and the like sent from the information processing apparatus 1. Can be heard by voice. From the management terminal 45, not only voice but also character data (not shown) can be input from a keyboard.

The present embodiment having the above configuration is characterized in that an index is set as reception information according to the connection mode of the communication terminal. That is, the threshold value is set according to the connection mode to the information processing apparatus 1 such as the telephone 4 connected to the telephone line 2, the Internet telephone 46 connected to the Internet, or the management terminal 45 directly connected to the information processing apparatus 1. I made it.

In the present embodiment, a case will be described as an example where voices are input from the telephone 4, the Internet telephone 46, and the management terminal 45, respectively.

First, a case where a voice is inputted from the telephone 4 connected to the telephone line 2 will be described as an example. Since the operation in the present embodiment is basically the same as that in the second embodiment, the common processing will be briefly described with reference to the flowcharts shown in FIGS.

When a call is made to the information processing apparatus 1 from the telephone 4 connected to the telephone line 2, the system management board 26 connected to the telephone line 2 receives the telephone tone (access request) (step 201). The "registered" name "" which is speech-synthesized by the speech synthesis circuit 39.
Please say Is transmitted (step 202). When the "name" (voice) from the voice input person corresponding to the voice response message is received (step 203), the next administrator authentication process is performed (step 204).

In FIG. 8, the voice recognition board 20
After extracting features from “name” (data) and generating a feature vector, distance calculation is performed on the standard vector of each registrant in the registrant data, and rearrangement is performed in ascending order of distance (step 221). For the registrant data sorted in ascending order, a threshold based on “access mode: telephone connected to telephone line” is set (step 222), and those less than the threshold are discarded (step 223).
For example, if the input is a voice input from the telephone 4 connected to the telephone line 2, the threshold for narrowing down the voice recognition is set to 5. In the case where the access mode is the telephone 4 connected to the telephone line (extension) 2, access from the office is assumed. However, since the telephone 4 can be easily used by a person other than the administrator, the security level is reduced. This is because it is better to make it relatively high and prioritize recognition accuracy over authentication efficiency.

If the input “name” (data) is included in the truncated registrant data of the voice data less than the threshold, the voice creator is authenticated as a registered manager. (Steps 224, 225). On the other hand, if it is not included in the registrant data after the truncation, it is determined that the voice input person is not a registered manager (step 22).
4,226).

The processing after step 205 in FIG. 7 is the same as that in the second embodiment, and a description thereof will be omitted.

Next, a case where a voice is input from the Internet telephone 46 connected to the Internet will be described as an example. Since the Internet telephone 46 has a voice conversion function, an example in which voice conversion is performed on the Internet telephone 46 side will be described based on the flowcharts shown in FIGS. 14 and 8.

When a call is made to the information processing apparatus 1 from the Internet telephone 46 connected to the Internet, a request is issued to the processor of the CPU board 13 or the CPU board 14 via the network board 47 of the information processing apparatus 1, and furthermore, the serial communication is performed. A connection request is issued to the system management board 26 via the board 19. The processor 31 of the system management board 26 receives this request via the serial communication circuit 34 (step 40).
1) The "response message" (data) is transmitted to the Internet telephone 46 connected to the Internet via the serial communication circuit 34, the serial communication board 19 of the information processing apparatus 1, and the network board 47.
This “response message” (data) is, for example, “Please say your registered“ name ”” (step 402).

In the Internet telephone 46, the "response message" (data) received via the LAN board 53 is converted into a "response message" (voice) by the sound circuit 54, so that the contents of the response message can be transmitted to the headphone set to the manager. 57 can be heard by voice.

[0102] The voice input person inputs a "name" (voice) from the Internet telephone 46 for authentication in response to the voice response message. The “name” (sound) is encoded into “name” (data) by the sound circuit 54, and is transmitted via the LAN board 53, the network board 47 of the information processing apparatus 1, and the serial communication board 19 to the system management board 26. Sent to. When the processor 31 of the system management board 26 receives the "name" (data) (step 403), the audio data memory 3
8 to the voice recognition board 20 via the serial communication circuit 34 and the serial communication board 19 of the information processing apparatus 1. Upon receiving the "name" (data), the voice recognition board 20 performs the following administrator authentication processing (step 40).
4).

In FIG. 8, the voice recognition board 20
After extracting features from “name” (data) and generating a feature vector, distance calculation is performed on the standard vector of each registrant in the registrant data, and rearrangement is performed in ascending order of distance (step 221). For the registrant data sorted in ascending order, a threshold based on “access mode: Internet telephone connected to the Internet” is set (step 222), and those less than the threshold are discarded (step 223). For example, if the input is a voice input from the Internet telephone 46 connected to the Internet, the threshold for narrowing down the voice recognition is set to 4. When the access mode is the Internet telephone 46 connected to the Internet, the security level is set to be relatively lower than that of the telephone 4 because the voice input person accesses the information processing apparatus 1 knowing the Internet address. This is because it is better to prioritize recognition efficiency over authentication accuracy.

If the input “name” (data) is included in the truncated registrant data of voice data less than the threshold value, the voice creator is authenticated as a registered manager. (Steps 224, 225). On the other hand, if it is not included in the registrant data after the truncation, it is determined that the voice input person is not a registered manager (step 22).
4,226).

Referring to FIG. 14, the voice recognition board 20
If it is determined in the administrator authentication process that the voice input person is not a registered administrator, the established line is disconnected (step 405). On the other hand, if the voice recognition board 20 determines that the voice input person is the registered manager in the manager authentication process (step 405), the serial communication board 19 and the serial communication circuit 34 of the system management board 26 are again used. Is transmitted to the audio data memory 38 via Here, the processor 31 is provided with the ROM 32 or the R
By executing a predetermined application stored in the AM 33, a predetermined “response message” (data) is transmitted to the Internet via the serial communication circuit 34, the serial communication board 19 of the information processing apparatus 1, and the network board 47. Connected internet phone 4
6 (step 406). The “response message” (data) is, for example, “please state the processing contents to access the information processing device.”

The Internet telephone 46 converts the “response message” (data) received via the LAN board 53 into a “response message” (voice) by the sound circuit 54, so that the contents of the response message can be transmitted to the headphone set by the administrator. 57 can be heard by voice.

In response to the voice response message, the voice input person inputs "state" (voice) from the Internet telephone 46 for authentication. This “state” (sound) is encoded into a “state” (data) by the sound circuit 54, and is transmitted to the system management board 26 via the LAN board 53, the network board 47 of the information processing apparatus 1, and the serial communication board 19. Sent to. When the processor 31 of the system management board 26 receives the “state” (data) (step 407), the processor 31 of the audio data memory 3
8 to the voice recognition board 20 via the serial communication circuit 34 and the serial communication board 19 of the information processing apparatus 1. The voice recognition board 20 converts the received “state” (data) into a “state” (character code) and sends it back to the system management board 26. The "state" (character code)
Are serial communication board 19 and serial communication circuit 34
Is stored in the audio data memory 38 via The processor 31 performs the same processing as in the first embodiment based on the "state" (character code) (step 40).
8) The processing result (data) is transmitted to the Internet telephone 4
6 (step 409).

In the Internet telephone 46, the processing result (data) received via the LAN board 53 is converted into a processing result (voice) by the sound circuit 54, so that the state of the information processing apparatus 1 can be transmitted from the headphone set to the manager. Can be heard. The “failure location, content” (voice) as the processing result is, for example, “AC / DC unit # 1, output voltage drop”.

As described above, the authentication and management processing of the administrator can be executed for the communication terminal having the voice conversion function without performing the voice conversion in the information processing apparatus 1.

Next, a case where a voice is input from the management terminal 45 directly connected to the system management board 26 of the information processing apparatus 1 will be described as an example. The management terminal 45 has substantially the same configuration as the Internet telephone 46, except that data is directly transmitted to and received from the system management board 26 without passing through the network board 47, so that the description is simplified. That is, the processing here is based on the flowcharts shown in FIGS.

When the management terminal 45 is connected to the information processing apparatus 1, the system management board 26 reads “registered”.
A response message (data) saying "Please say name" is transmitted (steps 401 and 402). When the "name" (data) from the voice input person corresponding to the response message is received (step 403), the next manager authentication process is performed (step 404).

In FIG. 8, the speech recognition board 20
After extracting features from “name” (data) and generating a feature vector, distance calculation is performed on the standard vector of each registrant in the registrant data, and rearrangement is performed in ascending order of distance (step 221). For the registrant data sorted in ascending order, a threshold based on “access mode: management terminal” is set (step 222), and those less than the threshold are discarded (step 223). For example, management terminal 4
If the input is a voice input from 5, the threshold for narrowing down the voice recognition is set to 3. When the access mode is the management terminal 45 directly connected to the information processing apparatus 1,
Since the access is from within the office, and the access at the time of maintenance is assumed compared to the case of the telephone 4 or the Internet telephone 46, the security level is made lower than that of the Internet telephone 46, and the recognition efficiency is given higher priority than the authentication accuracy. This is because it is better to do it.

If the input “name” (data) is included in the truncated registrant data of voice data less than the threshold, the voice creator is authenticated as a registered manager. (Steps 224, 225). On the other hand, if it is not included in the registrant data after the truncation, it is determined that the voice input person is not a registered manager (step 22).
4,226).

The processing after step 405 in FIG. 14 is the same as the processing in the above-described Internet telephone 46, and therefore the description is omitted.

As described above, according to the present embodiment, the telephone 4, the Internet telephone 46, and the management terminal 45 respectively make or connect to the information processing apparatus 1 to perform authentication based on the input voice. When the authentication function determines that the administrator is a registered voice, the administrator can obtain the status of the information processing apparatus 1 by voice. In the present embodiment, the threshold for narrowing down the authentication function by voice is changed according to the connection form such as telephone line, Internet or direct connection. Thereby, the recognition accuracy or the recognition efficiency can be selectively improved according to the connection mode.

In the present embodiment, the connection mode for setting the threshold value is described by taking as an example the case of a telephone line, the Internet, or a direct connection. However, the present invention is not limited to this.

Further, in the present embodiment, the state acquisition processing has been described as an example, but it goes without saying that the present invention can be applied to other management processing such as a configuration change processing.

In each of the above embodiments, the setting of a threshold value as an index according to the contents of processing or as information obtained at the time of reception according to an access route, an access mode, the number of times of input, etc. Although the description has been made, it is also possible to use a combination of conditions such as the content of the processing serving as a reference for setting the threshold. Further, in the present embodiment, for example, a threshold value is set in association with each item of management processing such as acquisition of a state or change of a configuration. However, when a plurality of conditions are used in combination, a predetermined expression is used. The threshold value may be determined by a calculation based on the threshold value.

Further, in each of the above embodiments, the threshold value is used as an index. However, the present invention can be applied to a case where the registrant data arranged in ascending order is narrowed down by the number from the top.

[0120]

According to the present invention, it is possible to confirm whether or not a voice input person is a registered manager by performing authentication on voice input from a communication terminal capable of voice input. Therefore, it is possible to easily authenticate the administrator without inputting a password or the like.

Further, an index for narrowing down the voice data of the administrator who has been registered in advance when authenticating the administrator can be set in accordance with the content of the process. The recognition accuracy or the recognition efficiency can be improved.

In addition, an index for narrowing down the pre-registered manager's voice data to be used at the time of authentication of the manager corresponds to the reception information such as the input path, the number of input information inputs, and the connection form of the communication terminal. So that it can be set
The recognition accuracy or the recognition efficiency can be improved according to the content of the processing.

[Brief description of the drawings]

FIG. 1 is a block diagram of the entire system to which an information processing apparatus management system according to a first embodiment of the present invention is applied.

FIG. 2 is a block diagram of an information processing apparatus according to the first embodiment.

FIG. 3 is a block diagram of a system management board according to the first embodiment;

FIG. 4 is a flowchart showing an operation in the first embodiment.

FIG. 5 is a flowchart showing an administrator authentication process according to the first embodiment.

FIG. 6 is a block diagram of an entire information processing apparatus management system according to a second embodiment of the present invention;

FIG. 7 is a flowchart showing an operation in the second embodiment.

FIG. 8 is a flowchart illustrating an administrator authentication process according to the second embodiment.

FIG. 9 is a block diagram of an entire information processing apparatus management system according to a third embodiment of the present invention;

FIG. 10 is a block diagram of an entire information processing apparatus management system according to a fourth embodiment of the present invention;

FIG. 11 is a block diagram of an information processing apparatus according to a fourth embodiment.

FIG. 12 is a block diagram of an Internet telephone according to a fourth embodiment.

FIG. 13 is a block diagram of a management terminal according to the fourth embodiment.

FIG. 14 is a flowchart showing an operation in the fourth embodiment.

FIG. 15 is a diagram showing a control device and a control method of a conventional computer security system.

[Explanation of symbols]

1 information processing device, 2 telephone line (extension), 3 LA
N, 4, 41, 42 telephone, 5 PBX, 6 telephone line (external line), 7 fax machine, 8, 45 management terminal, 9
Router, 11,48,60 system bus, 12,5
1,63 I / O bus, 13,14 CPU board, 1
5 memory board, 16 bus bridge board, 17,
18 SCSI control board, 19 serial communication board, 20 voice recognition board, 21, 22, SCSI bus, 23, 24, 56, 68 disk unit, 25
System management bus (JTAG bus), 26 system management board, 27 power supply unit, 28 fan unit, 31 processor, 32 ROM, 33 RAM,
34,65 serial communication circuit, 35 telephone line circuit,
36 voice detection circuit, 37 voice control circuit, 38 voice data memory, 39 voice synthesis circuit, 40 voice encoding circuit, 46 Internet telephone, 47 network board, 49, 61 CPU, 50, 62 memory, 5
2,64 SCSI controller, 53 LAN board, 5
4,66 sound circuit, 55,67 I / O bridge, 57,69 Headphone set.

Claims (9)

[Claims] An information processing apparatus management system that performs management processing of an information processing apparatus based on input information from an administrator who is authorized to manage the information processing apparatus and transmits a processing result as necessary. Voice recognition means for recognizing voice input from a communication terminal capable of voice input; and a voice input person being an administrator by comparing voice recognized by the voice recognition means with voice data of a pre-registered administrator. Voice authentication means for performing voice authentication as to whether or not the voice authentication means, and voice data selection means for narrowing voice data of the administrator to be compared by the voice authentication means from all registered voice data, The data selection means sets an index for narrowing down the audio data to be subjected to the comparison process according to the content of the management process based on the audio sent from the communication terminal. The information processing apparatus management system according to claim Rukoto. 2. An information processing apparatus management system which performs management processing of an information processing apparatus based on input information from an administrator who is authorized to manage the information processing apparatus and transmits a processing result as necessary. Voice recognition means for recognizing voice input from a communication terminal capable of voice input; and a voice input person being an administrator by comparing voice recognized by the voice recognition means with voice data of a pre-registered administrator. Voice authentication means for performing voice authentication as to whether or not the voice authentication means, and voice data selection means for narrowing voice data of the administrator to be compared by the voice authentication means from all registered voice data, The data selection means sets an index for narrowing down audio data to be subjected to the comparison processing based on the reception information of the input information. Location management system. 3. The information processing apparatus management system according to claim 2, wherein the voice data selecting unit sets the index according to an input path of the input information as the received information. 4. The voice data selecting means sets the narrowing level set as the index to be higher when the input information is input outside the local line than when input information is input inside the local line. 4. The information processing apparatus management system according to claim 3, wherein: 5. The information processing apparatus management according to claim 2, wherein the voice data selection means sets the index according to the number of times of input information from each of the communication terminals as the received information. system. 6. The information processing apparatus management system according to claim 2, wherein the voice data selection unit sets the index according to a connection mode of the communication terminal as the received information. 7. The voice data selecting means does not have a common feature when the voice data of the manager is arranged in an order having a feature common to the voice recognized by the voice recognition means. 3. The method according to claim 1, wherein a threshold value for discarding audio data is used as the index.
The information processing device management system according to any one of the above. 8. An information processing apparatus management system that performs management processing of an information processing apparatus based on input information from an administrator who is authorized to manage the information processing apparatus and transmits a result of the processing as necessary. A voice recognition step of recognizing processing content by voice input from a communication terminal capable of voice input; and index setting for setting an index for narrowing down voice data in accordance with management processing content by voice sent from the communication terminal. And comparing the voice data narrowed down based on the index among the voice data of the administrator registered in advance with the voice input from the communication terminal to determine whether the voice input person is the manager. A voice authentication step of performing authentication, and determining whether or not to execute a management process according to a result of the voice authentication. Management method. 9. An information processing apparatus management system which performs management processing of an information processing apparatus based on input information from an administrator who is authorized to manage the information processing apparatus and transmits a processing result as necessary. A voice recognition step of recognizing processing content based on voice input from a communication terminal capable of voice input; an index setting step of setting an index for narrowing down voice data based on reception information of the input information; A voice authentication step of performing voice authentication as to whether or not the voice input person is a manager by comparing voice data narrowed based on the index and voice input from the communication terminal among voice data of the user, And determining whether or not to execute the management process according to the result of the voice authentication.