JP6731631B2 - Cognitive function evaluation device, program - Google Patents

Description

The present invention, cognitive function evaluation device, about the program. The present invention is more particularly directed to cognitive evaluation equipment for evaluating the cognitive function of a subject by speech of the subject. The present invention also relates to a program that causes a computer to function as a cognitive function evaluation device, and a program that realizes a cognitive function evaluation method on a computer.

Conventionally, a technique has been proposed in which a prosody feature amount is extracted from voice data and the risk level of cognitive impairment for a speaker of the voice data is calculated from the prosody feature amount of the voice data (for example, refer to Patent Document 1). The cited document 1 describes a technique for calculating the risk of cognitive impairment based on a combination of a plurality of types of prosodic feature quantities extracted from voice data and weighting for each of the combined prosodic feature quantities. There is. In Cited Document 1, the prosody feature amount is a feature amount regarding a frequency component of a voice, a feature amount regarding a voice formant structure, a feature amount regarding a voice volume, a feature amount regarding an utterance speed, and a feature regarding a reaction time until answering a question. At least one of the amounts. Then, for each feature amount, a value is obtained for each of the collected voices.

JP, 2011-255106, A

In Patent Document 1, a plurality of types of prosody feature amounts are obtained from collected voices, but one type of prosody feature amount is represented by one type of value. Therefore, a relatively large number of prosody feature quantities are required.

The present invention shall be the object of the invention to provide a cognitive test system to facilitate the extraction of the feature amount for evaluating cognitive functions. A further object of the present invention is to provide a program that causes a computer to function as a cognitive function evaluation device, and a program that implements a cognitive function evaluation method on the computer.

The cognitive function evaluation device according to the present invention includes an acquisition unit, a feature extraction unit, and an estimation unit. The acquisition unit acquires the voice data of the target person in the utterance section. The feature extraction unit extracts a feature amount that reflects the cognitive function of the target person from the voice data in the utterance section. The estimation unit estimates the cognitive function of the subject based on the change in the feature amount. The feature amount is represented by one type of variable time series data that reflects prosody information in a predetermined extraction period of the voice data.

A program according to the present invention is a program for causing a computer to function as a cognitive function evaluation device.

Another program according to the present invention is a program for realizing a cognitive function evaluation method on a computer.

According to the configuration of the present invention, there is an advantage that it is easy to extract the feature amount for evaluating the cognitive function.

FIG. 1 is a block circuit diagram showing a configuration example of a cognitive function evaluation device. FIG. 2 is a waveform diagram illustrating a speech section in a configuration example of the cognitive function evaluation device. FIG. 3 is a graph showing changes in the feature amount in the configuration example of the cognitive function evaluation device. FIG. 4 is a graph showing a change in another feature amount in the configuration example of the cognitive function evaluation device. FIG. 5 is a block circuit diagram showing another configuration example of the cognitive function evaluation device. FIG. 6 is a block circuit diagram showing a mode of use of the cognitive function evaluation device.

The cognitive function evaluation device described below is configured to evaluate the cognitive function based on the utterance of a person. The cognitive function is evaluated using at least one of the pitch frequency and formant contained in the human voice. Therefore, a person's cognitive function can be evaluated without understanding the utterance content.

As shown in FIG. 1, the cognitive function evaluation apparatus 10 includes an acquisition unit 11 that acquires voice data of a subject, and a feature extraction unit 12 that extracts a feature amount from the voice data acquired by the acquisition unit 11. Further, the cognitive function evaluation device 10 includes an estimation unit 13 that estimates the cognitive function of the subject based on the change in the characteristic amount. The subject is mainly intended for the elderly, but may be a person suspected of having juvenile Alzheimer's disease. Elderly people who are mainly in need of monitoring, such as those who live in welfare facilities, those who use day service centers, those who live alone, or those who live in housing for the elderly with services. Assume

The acquisition unit 11 is configured to perform analog-to-digital conversion for converting a voice signal of a subject input from a microphone into voice data that is digital data, or a voice of the subject recorded in a state of being converted into digital data. A structure in which data is input is adopted. That is, the acquisition unit 11 may have either a configuration for converting an analog signal into a digital signal or a configuration for receiving a digital signal. Here, it is assumed that the former configuration is adopted. The microphone may be integrated with the cognitive function evaluation device 10.

By the way, when a person speaks, generally, there are a period in which voice is continuously generated and a period in which voice is stopped. The pause of the voice is determined by the feature extraction unit 12 based on the sound pressure and the time. That is, the feature extraction unit 12 detects a state in which no voice is generated based on the sound pressure, and when this state continues for a predetermined determination time, determines that the voice is in a silent section. The determination time is set to, for example, about 300 [ms] or more and 500 [ms] or less. The determination time can be shortened or extended depending on the target person. In addition, the feature extraction unit 12 detects a state in which voice is generated based on the sound pressure, and when this state continues without interposing a silent period, determines that the voice period is a continuous utterance period. To do. It is desirable that a time limit is set for the utterance section. The time limit is set to, for example, 15 [s] or more and 30 [s] or less.

As shown in FIG. 2, the feature extraction unit 12 uses the time-series data of the variable Va obtained for each predetermined extraction period Tx for one utterance section Ts in the voice data as the feature amount. The variable Va is obtained from at least one of the pitch frequency and the formant. The feature extraction unit 12 analyzes the spectrum of the voice data by performing a short-time Fourier transform using a short-time window function or a wavelet transform in order to obtain the pitch frequency and the formant. When the variable Va is obtained by using the formants, three types from the first formant to the third formant may be used, but it is also possible to use two types or less or four types or more.

As described above, the feature amount obtained by the feature extraction unit 12 is represented by time-series data of one type of variable Va that reflects prosody information for each extraction period Tx in the utterance section Ts. That is, the extraction period Tx is set such that the utterance section Ts includes a plurality of extraction periods Tx and the variable Va obtained for each extraction period Tx reflects the prosody information of the voice data. The extraction period Tx is set to, for example, about 0.5 [s] or more and 2 [s] or less. Here, the extraction period Tx is set to 1 [s].

By the way, the inventors of the present invention appropriately determine the length of the extraction period Tx, and appropriately determine the representative value of the pitch frequency or the formant in the extraction period Tx, so that the normal subject and the subject whose cognitive function is deteriorated. We obtained the knowledge that the voices of and can be distinguished. That is, in the utterance period Ts, the change in the pitch frequency or the representative value of the formant in the extraction period Tx is relatively small in the voice of the normal subject and is relatively large in the voice of the subject whose cognitive function is deteriorated. Knowledge was obtained.

Now, assume that the feature amount is obtained from the pitch frequency. In this case, the representative value of the pitch frequency in the extraction period Tx is selected from an average value, a median value, a mode value, a maximum value, a minimum value and the like. Desirably, the representative value in the extraction period Tx is selected from the average value, the median value, and the mode value of the pitch frequencies in the extraction period Tx. Further, the representative value is standardized so as to be represented by a numerical value in a predetermined range.

For example, the case where the pitch frequency is 500 [Hz] is set to “1”, and the feature extraction unit 12 sets “n” when the representative value in the extraction period Tx is n times 500 [Hz]. The variable Va of the extraction period Tx is set. In this case, when the representative value in the extraction period Tx is 1/n of 500 [Hz], the feature extraction unit 12 sets “−n” to the variable Va of the extraction period Tx. This variable Va is an example, and for example, when the pitch frequency is 300 [Hz], it is set to “0”, and the feature extraction unit 12 sets the representative value in the extraction period Tx to 300 [Hz]×2 ^n. In this case, “n” may be set to the variable Va of the extraction period Tx. In any case, the feature extraction unit 12 calculates a value obtained by normalizing the representative value of the pitch frequency in the extraction period Tx as the variable Va.

The feature extraction unit 12 can obtain the variation amount Va from a plurality of formants, and can obtain the variation amount Va from the pitch frequency and the formants. In this case, since a plurality of representative values are obtained for the extraction period Tx, the feature extraction unit 12 synthesizes the plurality of representative values and then determines a value obtained by normalizing the synthesized value as the variable Va. When synthesizing a plurality of representative values, it is desirable to perform weighting according to the representative value.

Since the feature extraction unit 12 obtains the variable Va for each extraction period Tx in the utterance section Ts, a plurality of variables Va are obtained in the utterance section Ts. The time-series data of the variable Va is the feature amount, and since the feature amount is obtained based on at least one of the pitch frequency and the formant, it reflects the prosodic information in the extraction period Tx of the voice data. In addition, since the variable Va is aggregated into one type of information, the feature amount is represented as time-series data of one type of information. Moreover, since the feature amount is standardized, it is represented by a numerical value within a predetermined range with respect to the actual voice data. The variable Va is normalized, for example, so that the closed interval [−10, 10] is set as the range. Further, the feature extracting unit 12 may round the variable Va obtained as described above to an integer value in order to represent the variable Va with an integer value.

It has been found that the feature amount obtained by the feature extraction unit 12 is relatively small when the cognitive function is normal, and is large when the cognitive function is deteriorated. Therefore, the estimation unit 13 estimates the cognitive function based on the change in the feature amount. For example, it is assumed that the feature amount as shown in FIG. 3 is obtained in one utterance section Ts. In the example shown in FIG. 3, the variable Va has changed in the range of 0 or more and 9 or less, and even if the period in which the variable Va is 0 is excluded, the difference of the variable Va is about 4. That is, in this example, it can be said that the change in the feature amount is relatively large.

The estimation unit 13 uses the magnitude of the difference (range) between the maximum value and the minimum value of the feature amount in the utterance section Ts as one method of evaluating the change in the feature amount. In the example shown in FIG. 3, the maximum value is 9 and the minimum value is 0, so the range is 9. When using the magnitude|size of a range for evaluation of a feature-value, the estimation part 13 should just set the threshold value with respect to a range, and when a range exceeds a threshold value, it should just be evaluated as a fall of a cognitive function. It is also possible to classify cognitive function into multiple stages based on the size of the range.

The method of evaluating the feature amount in the estimation unit 13 is not limited to the above example. For example, in the utterance section Ts, the estimation unit 13 evaluates whether or not the feature amount between the two extraction periods Tx in which the variable Va is 0 is unimodal, and when it is not unimodal, the cognitive function deteriorates. You may evaluate it as doing. In other words, it is evaluated that the cognitive function is deteriorated when the state where the variable Va is 0 is minimized between the two extraction periods Tx. Further, in this case, the estimation unit 13 may evaluate that the cognitive function is deteriorated when the size of the range exceeds a predetermined threshold, even if it is unimodal.

By the way, since the estimation unit 13 evaluates the cognitive function based on the change in the feature amount, it is easier to evaluate the cognitive function by emphasizing the change in the feature amount. Therefore, the characteristic amount is represented by time-series data in which a value obtained by adding a plurality of adjacent predetermined variable amounts Va among the time-series data of the variable Va obtained for each extraction period Tx in the utterance section Ts is used as the data of the temporary point. May be. For example, when the time series data of the variable Va is represented by V(1), V(2),..., V(i),..., D(2)=V(2)+V(1),. )=V(i)+V(i−1),..., The time series data D(1), D(2),..., D(i),. Here, i means the i-th extraction period Tx in the utterance section Ts.

When the time series data D(1), D(2),..., D(i),... Are obtained from the feature amount shown in FIG. 3, the new feature amount shown in FIG. 4 is obtained. As can be seen by comparing FIG. 3 with FIG. 4, the change in the feature amount shown in FIG. 4 is emphasized more than the feature amount shown in FIG. 3, and therefore the change in the feature amount can be easily evaluated. For example, the range of the characteristic amount shown in FIG. 3 is 9, whereas the range of the characteristic amount shown in FIG. 4 is 15. Further, in the feature amount shown in FIG. 4, compared with the feature amount shown in FIG. 3, the gradient between the two extraction periods Tx in which the variable amount Va is 0 is emphasized, and a change occurs. Assuming that the estimation unit 13 sets the threshold value to 10, when the feature amount as shown in FIG. 4 is obtained, the estimation unit 13 evaluates that the cognitive function of the subject is deteriorated.

In the above-described configuration example, whether or not the cognitive function is normal is determined based on the voice data of one utterance section Ts. In contrast to this configuration, the estimation unit 13 may employ a configuration that determines whether or not the cognitive function of the target person is normal by comparing the feature amounts obtained from the voice data of the utterance section Ts of multiple times. Good. That is, as shown in FIG. 5, the cognitive function evaluation device 10 includes the storage unit 14 that stores the feature amount in the utterance section Ts of a plurality of times. As a general rule, the cognitive function evaluation device 10 desirably acquires voice data every day. Of course, the frequency with which the cognitive function evaluation apparatus 10 acquires the voice data may be about once a week, or may be multiple times a day.

As described above, the storage unit 14 stores the plurality of feature amounts respectively obtained from the voice data of the plurality of utterance intervals Ts. The estimation unit 13 evaluates the degree of similarity of the plurality of feature amounts stored in the storage unit 14, and evaluates the cognitive function of the subject based on the degree of similarity of the feature amounts. In this case, it is desirable that the feature amount when the cognitive function of the subject is normal is stored in the storage unit 14. If the characteristic amount when the cognitive function of the subject is normal is stored in the storage unit 14, the estimating unit 13 determines the similarity between the characteristic amount obtained for each utterance section Ts and the normal characteristic amount. By evaluating, it is possible to estimate changes in the cognitive function of the subject.

The estimation unit 13 uses, for example, the difference between the average values of the variables Va in the feature amounts for the similarity evaluation, and determines that the cognitive function may be deteriorated when the difference exceeds a predetermined threshold. When the cognitive function is suspected to be deteriorated by the evaluation of the similarity of the characteristic amounts, the change of the characteristic amounts may be further used to evaluate whether or not the cognitive function is deteriorated.

The cognitive function evaluation device 10 described above includes a processor that operates according to a program. That is, the cognitive function evaluation device 10 includes a computer as a main hardware element. This type of processor is selected from an MPU (Micro-Processing Unit) which requires a separate memory, a microcomputer (Microcontroller) having a memory in a single device, and the like. The cognitive function evaluation device 10 can be configured by a terminal device managed by the subject, but as shown in FIGS. 1 and 5, the terminal device 21 managed by the subject inputs voice. It is desirable that the cognitive function evaluation device 10 used as the device of FIG. The computer server 100 may be configured by one computer, or may be configured so that a plurality of computers communicating via a computer network cooperate with each other and behave like one computer from the user. Further, the computer server 100 may be constructed as a cloud computing system.

The program may be provided in a state of being stored in a ROM (Read Only Memory) of the memory, or may be provided in a computer-readable recording medium such as an optical disk, an external storage device, or a memory card. Further, the program may be provided through a telecommunication line such as the Internet. The program provided through the storage medium or the electric communication line is stored in a rewritable nonvolatile memory.

The terminal device 21 managed by the target person is selected from a smartphone, a tablet terminal, a wearable computer, a personal computer, or the like. In the following, it is assumed that the target person carries a smartphone. The cognitive function evaluation device 10 is implemented in a computer server 100 that communicates with a smartphone through a telecommunication line such as the Internet. Moreover, in order to use the cognitive function evaluation apparatus 10, a configuration example in which an application program (so-called “app”) is executed in a smartphone is assumed.

In this configuration example, when the subject inputs a voice on the smartphone in which the application is activated, the voice data is handed over to the cognitive function evaluation device 10 realized by the computer server 100. It is desirable that the voice input to the smartphone be a natural voice during conversation rather than a voice that reads a specific sentence to evaluate the cognitive function. Therefore, it is desirable to use, as the voice input to the smartphone, a voice when making a call, a voice when using a web service in a natural language of voice, or the like.

The evaluation result of the cognitive function of the subject may be notified to the subject if the cognitive function of the subject is normal, but if the cognitive function of the subject has declined to the extent that it is considered to be mild cognitive impairment. Should inform a third party who can respond to a decline in cognitive function. Here, the cognitive function is considered to be within the range of cognitive impairment when the cognitive function is reduced to the extent that it is considered to be mild cognitive impairment. Further, in the following, the degree to which the cognitive function is mild cognitive impairment is referred to as a “sign level”.

Therefore, as shown in FIGS. 1 and 5, the cognitive function evaluation apparatus 10 is managed by a third party when the estimation unit 13 evaluates that the cognitive function of the target person has decreased to the predictive level. It is desirable to include the notification unit 15 that notifies the other device 22. The other device 22 is, for example, a terminal device managed by the subject's family, a terminal device managed by the subject's family doctor, a caregiver, or the like. This terminal device is selected from, for example, a smartphone, a tablet terminal, and a personal computer.

In this way, when the cognitive function evaluation apparatus 10 evaluates the cognitive function of the target person, and it is estimated that the cognitive function of the target person has fallen to the predictive level, the notification unit 15 notifies the third party. Is possible. By notifying a third party that the cognitive function of the subject has declined to a predictive level, early diagnosis of whether the subject has cognitive impairment or early treatment of cognitive impairment is possible. It leads to things to do.

In the configuration example described above, it is premised that the target person uses the terminal device 21, but when the cognitive function of the target person is deteriorated, there is a possibility that the terminal device 21 cannot be handled. Therefore, the cognitive function evaluation apparatus 10 may acquire voice data from a communication robot having a dialogue function. Generally, a communication robot has a function of simulating a dialogue with a person or a function of mediating a dialogue with another person. Therefore, the cognitive function evaluation apparatus 10 can acquire the voice data of the target person by cooperating with the communication robot.

An example of this configuration is shown in FIG. The basic configuration is the same as the configuration example shown in FIGS. 1 and 5, and a communication robot 23 is used instead of the terminal device 21. The communication robot 23 includes a microphone 231 and a speaker 232, and has a configuration that changes its appearance according to the content of conversation. Changing the appearance means changing the form, movement, light, color, and the like. In addition, the communication robot 23 includes a control device 230 that enables the communication robot 23 to interact with the target person by itself. Furthermore, the communication robot 23 includes a communication unit 233 that communicates with the computer server 100 via an electric communication line such as the Internet. That is, the communication robot 23 uses not only the information acquired by the control device 230, but also the information acquired from the computer server 100 through the communication unit 233 to interact with the target person.

By the communication robot 23 interacting with the target person in this way, the computer server 100 can acquire the voice data of the target person. Therefore, if the cognitive function evaluation apparatus 10 is realized by the computer server 100, the cognitive function evaluation apparatus 10 can evaluate the cognitive function of the target person based on the voice data acquired through the communication robot 23. ..

Since the communication robot 23 may not correspond to the target person in a one-to-one relationship, when evaluating the cognitive function of the target person using the voice data received from the communication robot 23, Need to be identified. Therefore, the cognitive function evaluation device 10 shown in FIG. 6 includes an authentication unit 16 that identifies the target person. The authentication unit 16 is assumed to have a configuration in which the target person is specified by the information of the voiceprint extracted from the voice data. However, when the communication robot 23 includes a camera for photographing the target person and the cognitive function evaluation device 10 acquires the image data captured by the camera, the authentication unit 16 performs face authentication based on the image data. It may be configured.

The cognitive function evaluation device 10 having the above-described configuration example includes an acquisition unit 11, a feature extraction unit 12, and an estimation unit 13. The acquisition unit 11 acquires voice data of the target person in the utterance section Ts. The feature extraction unit 12 extracts a feature amount that reflects the cognitive function of the target person from the voice data in the utterance section Ts. The estimation unit 13 estimates the cognitive function of the subject based on the change in the feature amount. The feature amount is represented by one type of variable time series data that reflects prosody information in a predetermined extraction period Tx of voice data.

According to this configuration, one kind of variable reflecting the prosody information is extracted for each extraction period Tx in the utterance section Ts, and the time-series data of this variable is used as the characteristic amount, so that the characteristic amount can be easily extracted. Further, since the feature amount is represented by one type of variable Va, it is possible to easily evaluate the cognitive function.

It is desirable that the estimation unit 13 evaluate the cognitive function of the subject based on at least one of the difference between the maximum value and the minimum value of the feature amount and the temporal change of the variable Va in the feature amount.

With this configuration, the estimation unit 13 can evaluate the cognitive function of the subject without performing complicated calculation using the feature amount.

It is desirable that the variable Va represents the pitch frequency extracted from the voice data for each extraction period Tx, as a numerical value within a predetermined range. In addition, the variable Va may be represented by a value in a predetermined range, which is a value obtained by combining a plurality of predetermined formants extracted from the voice data for each extraction period Tx. Alternatively, the variable Va represents a value obtained by combining a pitch frequency extracted from the voice data in each extraction period Tx and a plurality of predetermined formants extracted from the voice data in each extraction period Tx, with a numerical value in a predetermined range. Can also

That is, the variable Va is obtained from at least one of the pitch frequency and formant obtained from the voice data. Therefore, the variable Va is information that reflects the prosody of the voice data, and is represented by one type of variable Va. In addition, since the variable Va is standardized so as to be represented by a numerical value within a predetermined range, it is possible to suppress the influence of individual differences and perform an objective evaluation.

The characteristic amount is preferably represented by time-series data in which a value obtained by adding a plurality of adjacent predetermined variable amounts Va among the time-series data of the variable Va is used as data at a temporary point.

For example, when the time series data in which two adjacent variables Va are added in the time series data of the variable Va are arranged along the time axis as the feature amount, the change of the variable Va is emphasized. Therefore, the estimation unit 13 can easily evaluate the change in the feature amount, and can improve the accuracy of evaluation of the cognitive function.

Further, it is desirable that the cognitive function evaluation apparatus 10 includes a storage unit 14 that stores the feature amount in the utterance sections Ts of a plurality of times. In this case, it is preferable that the estimation unit 13 evaluate the degree of similarity of the feature amounts in the plurality of utterance sections Ts stored in the storage unit 14 and estimate the cognitive function of the subject according to the degree of similarity. ..

According to this configuration, assuming that the initial utterance section Ts is the normal utterance section Ts in the plurality of utterance sections Ts, when the degree of similarity of the feature amounts in the plurality of utterance sections Ts decreases. It is possible to estimate that the cognitive function of the subject has deteriorated. With this configuration, it is possible to evaluate the cognitive function of the subject while suppressing the influence of individual differences in the voice of the subject.

It is preferable that the cognitive function evaluation apparatus 10 further includes a notification unit 15 that notifies the other device 22 when the estimation unit 13 estimates the cognitive function of the subject as a range of cognitive impairment.

According to this configuration, when the cognitive function declines to the extent that it is estimated to be within the range of cognitive impairment, the other device 22 is notified, and a third party can be prompted to have the subject undergo an examination or treatment. It will be possible.

In the cognitive function evaluation method in this configuration example, a step of extracting a feature amount from the voice data of the subject in the utterance section Ts for each predetermined extraction period Tx, and estimating the cognitive function of the subject based on the change in the feature amount. And steps. The feature amount is represented by one type of variable time series data that reflects prosody information in a predetermined extraction period Tx of voice data.

According to this method, since the feature amount is represented by one type of variable Va, it is possible to easily evaluate the cognitive function.

The program in this configuration example is a program for causing a computer to function as the cognitive function evaluation device 10. Alternatively, the program in this configuration example is a program for realizing the cognitive function evaluation method on a computer.

The above-described embodiment is an example of the present invention. For this reason, the present invention is not limited to the above-described embodiment, and other than this embodiment, as long as it does not deviate from the technical idea of the present invention, various types according to the design etc. Of course, it can be changed.

10 cognitive function evaluation device 11 acquisition unit 12 feature extraction unit 13 estimation unit 14 storage unit 15 notification unit 22 other device Ts utterance section Tx extraction period Va variable

Claims (10)

An acquisition unit that acquires the voice data of the target person in the utterance section,
A feature extraction unit that extracts a feature amount that reflects the cognitive function of the subject from the voice data in the utterance section;
An estimation unit that estimates the cognitive function of the subject based on the change in the feature amount,
The cognitive function evaluation apparatus is characterized in that the feature amount is represented by one type of variable time series data that reflects prosody information in a predetermined extraction period of the voice data. The cognitive function according to claim 1, wherein the estimating unit evaluates the cognitive function of the subject based on at least one of a difference between the maximum value and the minimum value of the feature amount and a temporal change of the variable amount in the feature amount. Functional evaluation device. The cognitive function evaluation device according to claim 1 or 2, wherein the variable represents a pitch frequency extracted from the voice data for each extraction period by a numerical value in a predetermined range. The cognitive function evaluation apparatus according to claim 1, wherein the variable represents a value obtained by combining a plurality of predetermined formants extracted from the voice data for each extraction period, with a numerical value in a predetermined range. The variable is a value obtained by synthesizing a pitch frequency extracted from the voice data for each extraction period and a plurality of predetermined formants extracted from the voice data for each extraction period as a numerical value in a predetermined range. The cognitive function evaluation device according to claim 1 or 2. The feature quantity is represented by time-series data in which a value obtained by adding a plurality of adjacent predetermined variables among the time-series data of the variables is data of a temporary point. The cognitive function evaluation device described in 1. There are multiple utterance sections,
Further comprising a storage unit that stores the feature amount in the plurality of utterance intervals,
The estimation unit is
7. The cognitive function of the subject is estimated according to the degree of similarity of the feature amounts in the plurality of utterance sections stored in the storage unit, and the degree of similarity is estimated. The cognitive function evaluation device according to item 1. The cognitive function evaluation device according to claim 1, further comprising a notification unit that notifies another device when the estimation unit estimates the cognitive function of the target person as a range of cognitive impairment. A program for causing a computer to function as a cognitive function evaluation device,
The cognitive function evaluation device,
A feature extraction unit that extracts a feature amount for each predetermined extraction period from the voice data of the target person in the utterance section,
An estimation unit that estimates the cognitive function of the subject based on the change in the feature amount,
The feature amount is represented by one type of variable time series data that reflects prosody information in a predetermined extraction period of the voice data.
A program characterized by that. A program for realizing a cognitive function evaluation method on a computer,
The cognitive function evaluation method,
A step of extracting a feature amount for each predetermined extraction period from the voice data of the target person in the utterance section,
Estimating the cognitive function of the subject based on the change in the feature amount,
The feature amount is represented by one type of variable time series data that reflects prosody information in a predetermined extraction period of the voice data.
A program characterized by that.

Images