JP2002132283A - Medical control system - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a medical control system which is high in the function to identify whether a user previously making registration is himself or herself or not and is good in operability. SOLUTION: The registered words made correspondent to the prescribed function of medical apparatus are inputted by speech from a speech input means (S3a); the similarity of the registered words is calculated from the speech data (S3b); further the thresholds to make discrimination as to whether the user is himself or herself or not are set by each of the respective registered words by subtraction of suitable values from the similarity of the speech data or the like (S3c and S3d), by which the function to identify whether the user is himself or herself or not is made better than the case of the same thresholds and the good operability is obtained.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

[0001] 1. Field of the Invention [0002] The present invention relates to a medical control system for inputting voice.

[0002]

2. Description of the Related Art A medical control system having a function of performing voice recognition to reduce the burden on an operator and improve operability is disclosed in Japanese Patent Application Laid-Open Nos. 60-205593 and 7-124.
No. 173. In this technique, the operation may not operate as expected by the user due to erroneous recognition, and Japanese Patent Application Laid-Open No. H11-175095 has been proposed as an improvement measure.

[0003]

However, according to this technique, even when it is desired to change the operation or setting of the device only with the words of a specific person, for example, the person who corresponds to the surgeon in the operation scene, the device can be described in a language other than the surgeon. There was a problem that would work. Also, in the conventional voice authentication, since the setting of the criterion (threshold) for identification is constant (the same value) for a plurality of registered words, it is easy to determine the identity depending on the difference between the authentication words. However, there are some which are easily judged to be others, and even the registered person has a large variation in the judgment result due to the difference in words, which is inconvenient and troublesome for some users.

As described above, in the conventional voice authentication, since the setting of the criterion (threshold) for identification of a person is constant (the same value) in a plurality of registered words, it is easy to determine that the person is the person, Some were easy to judge, making it inconvenient for users.

(Object of the Invention) The present invention has been made in view of the above points, and provides a user-friendly medical control system having a high identification function of whether or not a user who has registered in advance is high. The purpose is to do. It is another object of the present invention to provide a medical control system capable of performing more accurate voice verification.

[0006]

A voice input means for inputting voice and a vocabulary set corresponding to a function of a medical device are previously input from the voice input means and registered as voice data of the vocabulary. A voice registration unit capable of registering voice data for each vocabulary; and a voice of the vocabulary input by the voice input and voice data of the same vocabulary registered in the voice registration unit are the same person. Threshold setting means for setting a threshold value for determining whether or not the voice data is registered, for each of the voice data registered in the voice registration means; and a threshold value set by the threshold setting means. By means of voice matching means for performing personal identification as to whether the voice of the vocabulary input from the voice registration means and the voice data of this vocabulary registered in the voice registration means belong to the same person, To As a result, the user inputs the voice of the corresponding vocabulary in order to operate or set the function of the desired medical device, and the voice data for the vocabulary is the same as the voice data of the vocabulary registered in advance. Since a threshold value for determining whether or not there is a word can be set for each vocabulary (voice data), the function of discriminating from others is high, and the usability is improved.

[0007]

Embodiments of the present invention will be described below with reference to the drawings. (First Embodiment) FIGS. 1 to 7 relate to a first embodiment of the present invention, and FIG. 1 shows a surgical operation system according to a first embodiment for performing a surgical operation under observation of an endoscope. FIG. 2 shows a schematic configuration of a system controller, and FIG.
Shows a specific example of a registered word, FIG. 4 shows a display example by the centralized control device when performing word registration, FIG. 5 shows a display example by the centralized control device when setting the threshold value, and FIG. 7 shows a display example by the centralized control device when setting the threshold value, and FIG. 7 shows a flowchart of the operation content.

As shown in FIG. 1, in a surgical operation system 1 for performing a surgical operation using an endoscope (as a medical control system), there are many types of peripheral devices to be used, and various medical devices are provided in one cart. Therefore, while a plurality of medical devices are separately mounted on several carts 10 and 20, these carts are collected at almost one place to improve operability.

For example, the first cart 10 has a monitor 11, a centralized display panel 12, a first TV camera 13a, a first light source 14a, a second TV camera 13b, and a second light source 14.
b, system controller 15, video mixer 16, V
A TR 17, a distributor 18, a communication connector 19, etc. are mounted, and a monitor 21, a high-frequency ablation device (electric scalpel) 22, an insufflation device 23, a CO2 cylinder 24, a suction bottle 25, a distributor 26, communication connector 2
7 etc. are mounted.

[0010] Then, various medical devices are provided in the first cart 1.
In the 0 and the second carts 20, they are electrically connected to distributors 18 and 26 provided in the respective carts 10 and 20 via a communication cable (not shown). Also, the first
The cart 10 and the second cart 20 are electrically connected via a universal cable 5 in which a communication cable is provided. Further, the first cart 10 and the second cart 20 and the peripheral device controller 30 are electrically connected via a universal cord 6 having a communication cable therein.

The peripheral device controller 30 has a first
The setting switches that need to be frequently used among the medical devices mounted on the cart 10 and the second cart 20 are integrated into the central control device 7 which is the central control input means.

That is, among the switches related to the setting of each medical device, only those which are frequently used are limited and collected in the central control device 7, so that the output section which controls the irradiation of energy to the human body can be accurately performed. In order to achieve this, an output control unit is provided separately.

FIG. 2 is a block diagram showing the configuration of the system controller 15. The system controller 15 includes the voice input means 31 and the controlled device (13, 14, 17, 2).
2, 23).

The inside of the system controller 15 judges whether the speech input from the speech input means 31 is recognized by the speech recognition means 32 and the speech input from the speech input means 31 match those of a person registered in advance. A voice collating unit 33, an information processing unit 34 for processing data from the voice recognizing unit 32 and the voice collating unit 33, a command generating unit 35 for sending a command to each controlled device, and a device for operating and setting medical devices It comprises a recording means 36 for recording voice data such as when registering voice data of a registered word (vocabulary) assigned to each function, and an arithmetic processing unit 37 for calculating a threshold value. Here, the voice input means 31 is, for example,
A pin microphone, a headset microphone, and the like can be considered.

FIG. 3 shows registered words, in which the registered words and the similarity in that case are displayed. FIG.
(A) shows the central control device 7 at the time of word registration,
When the “A” dialog is pressed in FIG. 4A to register the electric knife, the state shown in FIG. 4B is obtained. And
After registering words up to the dialog of "I", a threshold value is set. FIG. 5A shows a screen displayed by the central control device 7 when the threshold value is set. When the setting of the threshold is completed, the display of FIG. 5B is displayed.

In the present embodiment, as described later in the description of the operation of FIG. 7, the user himself / herself performs a process of pre-registering voice data of a plurality of registered words associated with predetermined functions in the recording means 36. After that, in order to execute a desired function, when a speech input of a registered word corresponding to the function is performed, identification (determination) of whether or not the speech input of the registered word is a pre-registered user is performed. Reference value (threshold value)
Can be set for each registered word, so that it is possible to efficiently identify whether or not the user is the reference value user (compared to the case where it cannot be set for each registered word). ing.

Next, the operation of the surgical operation system 1 (mainly for voice input such as word registration by voice) will be described with reference to FIG. An example will be described in which the setting of the electric scalpel is changed in the surgical operation system 1 of FIG.
When the power switch is turned on and the operation starts, first, it is determined in step S1 whether word registration is necessary (whether word registration or change).

If word registration is selected, step S2
If word registration has already been performed and word registration has not been performed and the registration is not selected, the process jumps to step S5. In the case of performing word registration, the registrant sequentially registers words serving as keywords associated with a predetermined function such as “Electric knife” shown in FIG. For example, the word “Electric scalpel” is a voice command of a function for setting a mode for operating and setting the electric scalpel. At this time, FIG.
An input operation is performed according to the guidance of the central control device 7 shown in FIG.

When registering the word "Electric scalpel", refer to FIG.
(A) When the dialog of "A" is pressed and a voice (the word "electric scalpel") is generated toward the voice input means 31 within three seconds, an analog-to-digital converter (not shown) in the recording means 36 outputs the voice. The analog input voice input from the input means 31 is converted into digital information, and this digital information is stored in the data area of the recording means 36 as shown in FIG.
(Column of) • Recording is performed as voice data of “electric scalpel” in the portion (shown by row) of a.

FIG. 4B shows a state in which the voice of the registrant "Electric scalpel" is registered. This operation is repeated until registration of a required number of keywords is completed (in the present embodiment, a maximum of nine keywords are set).

When the word registration is completed, a process of setting an optimum threshold value for each word in step S3 is performed. In this process, the registrant inputs the registered word by voice in step S3a, calculates the degree of similarity by the voice collating means 33 in step S3b, and in step S3c, the arithmetic processing unit 37 determines whether or not the word is registered by the voice. A threshold value (which is a reference value for determination) is calculated, and in step S3d, it is determined whether or not the threshold setting has been completed for all the registered words and all registered words. , Steps S3a to S3d.

More specifically, the threshold value of the first registered word “Electric scalpel” can be set by pressing the “Electric scalpel” dialog with the word registered as shown in FIG. During the second, a voice (the word “electric scalpel”) is uttered toward the voice input means 31 (S3a), and the “input voice” is output.
(Similarity calculation) is performed by the voice matching means 33 to calculate the similarity of the input voice to the registered word, and the result is expressed as a percentage (S3b).

In the present embodiment, the similarity calculation as described above is performed once, but may be performed two or more times for one word. Then, in the arithmetic processing unit 37, the voice collating means 3
Using the similarity calculated in step 3, a threshold value is calculated for each word as follows, and stored in the recording unit 36 (S
3c).

If the similarity is calculated only once by the voice matching means 33, the threshold value of the word is obtained by subtracting an arbitrary value from the lowest value of the similarity. It is assumed that a threshold value when the similarity is calculated a plurality of times by the voice matching unit 33 is obtained by subtracting an arbitrary value from the lowest similarity.

FIG. 6 shows an arbitrary value setting screen when an arbitrary value is subtracted from the lowest similarity to set a threshold value. As shown in FIG. 6, a value (minus value) to be subtracted from the similarity can be freely determined for each word. Then, the determined subtracted value is subtracted from the similarity to determine a threshold value, and the threshold value is displayed in the display box.

In FIG. 6, if the same value is to be subtracted for all the words, the number to be subtracted is determined, and the dialog box labeled ALL is pressed. At this time, ALL is displayed in the threshold value display box.

The threshold value “75” of “Electric scalpel” calculated by the above method is reflected in the “Electric scalpel” dialog as shown in FIG. "And the threshold" 75 "are displayed alternately.

In this embodiment, since the threshold value can be set for each registered word, a higher identification (determination) function is provided as compared with the case where the same threshold value is set (the conventional example). This makes it possible to more easily perform system control by voice input, that is, to realize a device that is easy for the user to use.

To explain this situation more specifically, even if the user himself / herself, the similarity of the word to be registered is generally different for each word to be registered. When a threshold value is set, a common threshold value must be set, which is relatively high for registered words having low similarity and relatively low for registered words having high similarity.

In this case, for example, when a registered word having a low similarity is input by voice, the threshold value is relatively high even if the user inputs voice, so that the registered word is quite the user's own. Therefore, it is expected that the same registered word is repeatedly input by voice. Conversely, a registered word having a high degree of similarity may be identified as the user even if another person other than the user inputs voice.

On the other hand, in the present embodiment, a threshold value can be set in accordance with the similarity value of the registered word by the user's own voice input. Can be identified efficiently.

When the threshold setting in step S3 is completed for the registered word, a screen as shown in FIG.
In (B), not only one registered word but threshold values are set for all registered words), and the completion of threshold setting is transmitted to the operator. This threshold setting is performed for all registered words. As described above, the threshold value can be arbitrarily set for each registered word by the above-described threshold value setting method.

Next, voice input for operating / setting the (controlled) device in step S5 is performed. In this embodiment, since the setting of “electric scalpel” is performed, the operator utters “electric scalpel” to the voice input unit 31. Then, the input voice data is subjected to voice collation and voice recognition processing in the next step S6.

The input voice data of the voice “Electric scalpel” is sent to the voice recognition means 32 for use in voice recognition (step S6a), and is sent to the voice verification means 33 for voice verification (voice verification). Authentication (step S6b).

That is, the arithmetic processing section 37 sends the recognition result of “electric scalpel” to the information processing section 34. Subsequently, the information processing unit 34 receives the recognition result of the “electric scalpel” and sends a signal so that the voice collating unit 33 designates one registered word “electric scalpel” to be collated with the input voice. Send to

In the next step S6c, the similarity between the voice data of the input voice "Electric scalpel" sent to the voice collating means 33 and the voice data of "Electric scalpel" stored in the recording means 36 Is calculated by the voice matching means 33.
Then, it is determined whether or not the input voice data has exceeded the threshold value of the registered word recognized by the voice recognition means 32. For example, as a result of the voice verification by the voice verification unit 33, when the similarity to the registered voice data “Electric scalpel” is “80” and exceeds the threshold value “75” of the registered voice data “Electric scalpel”, the voice input The person is authenticated (identified).

On the other hand, if the input voice does not exceed the threshold, the process returns to step S5, and voice input for operating and setting the apparatus is performed again. In this case, if none of the thresholds set for each of the registered words has been exceeded, the input voice is determined to be other than the word registrant, and no device setting or the like is performed. In the present embodiment, the “voice recognition processing” and the “voice authentication (collation) processing” are performed serially, but may be performed in parallel by respective means.

If the input voice is authenticated by the voice collating means 33, the determination information that the input voice is the user is sent to the information processing section 34. The information processing unit 34 to which the determination information that the input voice is the person is sent,
The content recognized by the voice recognition means 32, in this case, the fact that "Electric scalpel" has been selected (recognition result) is sent to the command generator 35 for transmitting the controlled device (Electric scalpel 22) (step S6d).

The command generator 35 sends a command to select the electric knife to the controlled device (electric knife 22) (step S6e). Here, an electric scalpel (high-frequency ablation device) 22
Preparations for operations and setting changes are completed.

Subsequently, when the output of the electric scalpel 22 is increased, the operator utters “output” to the voice input means 31. If it is determined that the word belongs to the person who registered the word in advance, the electric scalpel 22 associated with the word enters the output change mode. "To operate the controlled device and change the setting (step S).
7).

In the present embodiment, word registration (step S2), threshold setting (step S3), voice recognition
Although the collation (step S6) is performed continuously, it may be performed individually instead of continuously.

The previously registered audio data may be read out from the recording means 36 and used. Also, a part of the registered voice data can be changed. The registration data is stored for each operator.

This embodiment has the following effects. By providing a judgment reference value (threshold value) for identification of each person for each word registration (voice), the function of identifying the individual with another person can be enhanced, and therefore, the operation and setting of the controlled device can be efficiently performed by voice input. This makes it easier for a user, such as an operator performing a treatment for treatment with a medical device, to be used by a user.

(Second Embodiment) Next, a second embodiment of the present invention will be described with reference to FIGS. Since the present embodiment is similar to the first embodiment, only different parts will be described.

In the present embodiment, the word necessary for operating the controlled device is not determined arbitrarily by the operator such as the operator, but is determined in advance by the manufacturer. Registration is performed in the same manner as in the first embodiment using the centralized control device 7 on which nine words specified by the manufacturer are displayed in advance.

The operation of the present embodiment will be described below with reference to the flowchart of FIG. The flowchart of FIG. 9 is different from the flowchart of FIG. 7 in that the voice collation / voice recognition processing in step S3 is partially different from that in step S3 ′, and the other processing is the same.

When the power switch is turned on and the operation starts, it is first determined in step S1 whether word registration is necessary (whether word registration or change). When the word registration is selected, the word registration in step S2 is performed, and when the word registration is already performed and no registration is selected, the process jumps to step S5. When the word registration is performed, the voice data of the user is recorded by the central control device 7 in which nine words specified by the manufacturer are displayed in advance as shown in FIG.
This is done by recording in Then, the next step S
Proceed to 3 '.

In the present embodiment, the number of words registered on the maker side is nine. However, the number is not limited to nine, and an arbitrary plural number may be used. FIG. 10 shows that the recording means 36 stores the contents of "voice data registered by the manufacturer" and "voice data registered by the user (operator)". In the present embodiment, the number of unspecified persons registered on the maker side is three, but the number may be arbitrary.

Although not shown in FIG. 10, in addition to the “electric scalpel”, “output” and “insufflation device”, “up”, “down”, “bipolar”, “monopolar”, “light source” and “insufflation device” Voice data "camera" is similarly recorded. Next, in the process of setting the optimum threshold value for each word in step S3 ', the registrant inputs the registered word by voice (step S3a), performs voice matching on the input voice,
The degree of similarity of the input voice to the registered word is expressed as a percentage and stored in the recording means 36 (step S3e).
It is determined whether or not the similarity is obtained (step S3f),
Thereafter, a threshold setting calculation is performed (step S3g), and a threshold is set for the registered word (step S3).
h) Further, the threshold value setting is repeated so as to be performed for all nine registered words (step S3i).

In the processing for setting the optimum threshold value for each word, for example, a case where the threshold value for the word "insufflation device" is set will be described. As shown in FIG. 11A, the centralized control device 7 is on a threshold setting screen, and the operator presses the “insufflation device” dialog box. "Device" and speak toward the voice input means 31 (step S3a).

In FIG. 10, four voice data, three voice data registered by the maker and three voice data registered by the user (operator), respectively, are input in step S3a. The voice data is collated with each other, the similarity is calculated for each, and the result is recorded in the recording means 36 (step S3e).

This similarity calculation is performed ten times. That is,
10 times for "Insufflation Device" registered by A, 10 times for "Insufflation Device" registered by B, 10 times for "Insufflation Device" registered by C, and The user (Ms. J) performs collation 10 times against the registered “pneumoperitoneum device” to calculate the similarity (step S3f). In this embodiment, the similarity calculation is performed ten times.
Times or any number of times.

FIG. 12 shows the information recorded in the recording means 36.
It shows the degree of similarity of the input speech "insufflation device" to the registered speech data "insufflation device" of four persons A, B, C and J. The similarity shown in FIG. 12 is sent from the recording unit 36 to the arithmetic processing unit 37. Then, step S3g,
The threshold setting of S3h is calculated, and the setting of the threshold for the registered word is completed.

As shown in FIG. 13, "distribution of similarity to another person registration" and "distribution of similarity to personal registration" overlap (FIG. 13A) and non-overlapping (FIG. 113B). )).

In any case, when the threshold value is “a”, the largest value of “similarity to another person registration” is b, and the smallest value of “similarity to another person registration” is b, (a +
b) / 2. For example, in FIG. 13B, the threshold value of the “pneumoperitoneum device” is (71 + 80) / 2 = 7
It is set to 5.5%.

When the similarity calculation is performed only once, the largest value among the three "similarities to other person registration (three)" is set to a, and the "one-person registration" calculated only once is used. The above calculation is performed with the value of “similarity to b” as b, and the threshold value is calculated. As described above, the threshold is set for the registered words, and it is determined in step S3i whether the threshold has been set for all the registered words. If not all, the process returns to step S3a and the same processing is repeated.

When the threshold setting for all the registered words is completed, the central controller 7 displays the threshold with the fractional part truncated, as shown in FIG. 11B, and the threshold setting is completed. (Step S4). Then, step S
The voice input processing of the operation / setting of the (controlled) device 5 is performed, and the voice collation / voice recognition of step S6 is performed.
Then, in step S7 (after the controlled device executes the content input by voice in step S5,
Will return to.

In this embodiment, it is possible to set the threshold value between “the similarity of oneself” and “the similarity of another person”. In addition, the threshold value can be calculated and set from “similarity of oneself” and “similarity of another person”.

This embodiment has the following effects. By calculating the threshold value of the voice authentication from the “similarity to other person registration” and the “similarity to personal registration”, more accurate voice authentication can be performed.

With this configuration, personal identification is performed by performing voice verification using words (voices) input by voice from the voice input means 31, and personal identification is further performed for each registered word (voice). By setting the determination reference value, it is possible to improve the personal authentication rate and to provide a user-friendly (by voice recognition) system control device.

[Supplementary Notes] Voice input means for inputting voice, voice registration means for registering a plurality of different voices for the same person, voice verification means for performing personal identification with voice input by the voice input means, and the voice verification means A medical control system provided with a recording unit for storing comparison data to be used, comprising a calculation processing unit capable of setting an arbitrary threshold value for each registered voice data. Control system.

[0062] 2. 2. The medical control system according to claim 1, wherein the threshold value calculated by the arithmetic processing unit is less than the similarity of the person. 3. 2. The medical control system according to claim 1, wherein the threshold value calculated by the arithmetic processing unit is a value between the similarity of the person and the similarity of another person. 4. 2. The medical device according to claim 1, wherein the threshold value calculated by the arithmetic processing unit is a value obtained by dividing a sum of a minimum value of the person's ex-post facto and a maximum value of similarity of another person by two. Control system.

[0063]

As described above, according to the present invention, the voice input means for inputting voice and the vocabulary set in advance corresponding to the function of the medical device are inputted from the voice input means in advance. Voice registration means for registering voice data for each vocabulary, and a voice of the vocabulary input by the voice input and a vocabulary identical to the vocabulary registered in the voice registration means Threshold setting means for setting a threshold for determining whether or not the voice data of the same person belongs to each of the voice data registered in the voice registration means; and the threshold setting means. The identification of whether or not the voice of the vocabulary input from the voice registration means and the voice data of this vocabulary registered in the voice registration means belong to the same person, based on the threshold value set in Row And voice collating means, for inputting the voice of the corresponding vocabulary in order to operate or set a desired function of the medical device, and voice data of the vocabulary in which voice data for the vocabulary is registered in advance. Vocabulary for each vocabulary (voice data)
Since it can be set for each user, the function of discriminating from others is high, and usability can be improved.

[Brief description of the drawings]

FIG. 1 is a schematic configuration diagram of a surgical operation system according to a first embodiment of the present invention.

FIG. 2 is a block diagram showing a configuration of a system controller.

FIG. 3 is a diagram showing a specific example of a registered word.

FIG. 4 is a diagram showing a display example by a centralized control device when performing word registration.

FIG. 5 is a diagram showing a display example by the central control device when setting a threshold value.

FIG. 6 is a view showing a display example by the central control device when setting an arbitrary threshold value.

FIG. 7 is a flowchart showing the operation contents.

FIG. 8 is a diagram illustrating a display example by a centralized control device when performing word registration according to the second embodiment of the present invention.

FIG. 9 is a flowchart showing operation contents.

FIG. 10 is a diagram showing that audio data of manufacturer registration and user registration is recorded in the recording means.

FIG. 11 is a diagram showing a display example by the central control device when setting a threshold value.

FIG. 12 is a diagram illustrating an example of the similarity between registered voice data of four persons.

FIG. 13 is a view showing a representative distribution type of the similarity by another person registration and the similarity by personal registration.

[Explanation of symbols]

 DESCRIPTION OF SYMBOLS 1 ... Surgical operation system 7 ... Centralized control device 10, 20 ... Cart 11 and 21 ... Monitor 12 ... Centralized display panel 13a, 13b ... TV camera 14a, 14b ... Light source 15 ... System controller 16 ... Video mixer 17 ... VTR 18, 26 ... Distributor 22 ... Electric knife 23 ... Insufflation device 19,27 ... Communication connector 30 ... Peripheral device controller 31 ... Speech input means 32 ... Speech recognition means 33 ... Speech verification means 34 ... Information processing unit 35 ... Command generation unit 36 ... Recording means 37 ... Calculation processing unit

 ────────────────────────────────────────────────── ─── Continuing on the front page (72) Inventor Junichi Onishi 2-43-2 Hatagaya, Shibuya-ku, Tokyo F-term in Olympus Optical Co., Ltd. 4C038 VA07 VB40 5D015 AA03 KK01 LL03

Claims (3)

[Claims] 1. A voice input unit for inputting voice, and a vocabulary set corresponding to a function of a medical device are previously input from the voice input unit and registered as voice data of the vocabulary. Voice registration means capable of registering voice data, and whether the voice of the vocabulary input by the voice input and the voice data of the same vocabulary registered in the voice registration means belong to the same person Threshold value setting means capable of setting a threshold value for determining whether or not the voice data is registered for each of the voice data registered in the voice registration means; and Voice matching means for identifying whether the voice of the vocabulary input from the registration means and the voice data of the vocabulary registered in the voice registration means belong to the same person. To Medical control system to a butterfly. 2. A voice input unit for inputting voice, and a vocabulary set corresponding to a function of a medical device are previously input from the voice input unit and registered as voice data of the vocabulary. Voice registration means capable of registering voice data; similarity calculation means for calculating the similarity between the voice of the vocabulary input by the voice input and the voice data of the vocabulary registered in advance in the voice registration means; The same person as the voice data of the vocabulary registered in the voice registration means calculates the similarity between the vocabulary and the voice input from the voice input means by the similarity calculation means,
Threshold setting means for setting in advance a threshold value for each of the voice data registered in the voice registration means based on the calculated similarity; voice of the vocabulary input from the voice input means and the voice Voice collating means for performing personal identification as to whether or not the voice data of the vocabulary registered in the register means belongs to the same person, based on a threshold value set by the threshold value setting means. A medical control system. 3. An operator setting means for limiting an operator by voice input to a single specific operator; a voice input means for inputting voice; and a vocabulary set corresponding to a function of the medical device. A voice registration unit that is input from a voice input unit and registered as voice data of the vocabulary, and is capable of registering voice data for each vocabulary; and a voice of the vocabulary input by the voice input and the voice registration unit in advance. A similarity calculating means for calculating a similarity of the voice data of the registered vocabulary; a voice of the vocabulary input by the specific operator by the voice inputting means; and the voice registered in the voice registering means in advance. Of the vocabulary voice data, the similarity between the specific operator's voice data of this vocabulary and the similarity of another person different from the specific operator with this vocabulary's voice data are calculated by the similarity calculation. A threshold setting unit that can set a threshold value in advance for each of the voice data registered in the voice registration unit based on each of the calculated degrees of similarity; The identification of whether or not the voice of the vocabulary and the voice data of the specific operator of the vocabulary registered in the voice registration means belong to the same person is set by the threshold value setting means. A medical control system comprising: a voice collating unit that performs a threshold value.