JP2005257905A - Cat's emotion discriminating device based on audio feature analysis of mewing - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide an emotion discriminating device which objectively discriminates the concrete emotion of a cat from mewing of the cat and to provide the emotion discriminating device whose discriminant result can be reflected on a game. <P>SOLUTION: The emotion discriminating device is equipped with; a conversion means 2 which converts the mewing of the cat into an electric audio signal; an input audio pattern extracting means 6 which extracts the feature of a relational map between a period and a frequency component of the audio signal as an input audio pattern; a reference audio pattern storage means 7 classified by emotion in which reference audio patterns classified by emotion which express for every various emotion features of relational maps between periods and frequency components in which the cat expresses characteristically various emotion as the mewing are stored; a comparison means 8 which compares the input audio pattern with the reference audio patterns classified by emotion; and a control means 10 which determines emotion whose correlation is the highest to the input audio pattern by the comparison and which displays the determined results on a picture display means 1. <P>COPYRIGHT: (C)2005,JPO&NCIPI

Description

  The present invention relates to an emotion discriminating device based on phonetic feature analysis, and more particularly to a cat emotion discriminating device based on voice feature analysis of crying.

  Conventionally, animals are often kept as pets that heal human hearts. In particular, dogs and cats were always near their owners, who judged their emotions based on animal behavior and crying, and determined what the animals are currently requesting and responded.

  Therefore, it is not an exaggeration to say that establishing communication with animals is a long-standing human dream, and various attempts have been made, and translation devices that translate the will of animals and the like have been proposed (for example, Patent Document 1). ).

This translation device receives voices uttered by animals such as pets and livestock, and converts them into audio signals, receives the movements of the animals as videos and converts them into video signals, and converts these audio signals and video signals into animal behavioral science etc. The will of the animal is translated by comparing the voice and motion data analyzed in advance.
Japanese Patent Laid-Open No. 10-3479

  The problem to be solved by the present invention is that, according to the translation device described above, it is considered possible to translate the animal's will based on the animal's voice and movement. Since specific voice and action data corresponding to emotions are not disclosed, after clearly grasping the relationship between the specific emotion of the cat and the singing characteristic of the cat with that emotion, It is not possible to objectively discriminate the cat's emotions based on the voice feature analysis based on the comparison between the cat's cry and the reference voice pattern. It was the point that the emotion of the cat could not be discriminated objectively based on the cry.

  The present invention provides an emotion discriminating apparatus that solves the above problems and objectively discriminates a specific cat's emotion from the cat's cry, and that can reflect the discrimination result in the game. The purpose is to provide.

  In order to solve the above problems, the cat emotion discrimination device based on the voice feature analysis of the cat's cry according to the present invention compares the input cat voice with the emotion-specific reference voice pattern, and displays the corresponding emotion discrimination result. An apparatus for discriminating an emotion of a cat displayed with images and sounds, characterized in that a conversion means for converting a cat's cry into an electric sound signal, and a feature of a relationship map between the time and frequency components of the sound signal are input sound patterns And an input voice pattern extracting means for extracting the emotion, and a reference voice pattern for each emotion representing the characteristics of the relationship map between the time and frequency components of the voice that the cat expresses various emotions characteristically as a cry Emotion-based reference voice pattern storage means, comparison means for comparing the input voice pattern with the emotion-specific reference voice pattern, and emotions having the highest correlation with the input voice pattern based on the comparison. In the cat emotion discrimination device based on the voice feature analysis of the cat's cry, comprising a control means for determining and displaying the determination result on the image display means, the emotion-based reference voice pattern is expressed as satisfaction emotion, courtship emotion, Intimate emotions, self-expression emotions, joy emotions, demand emotions, and for satisfaction, components are distributed in pulses at intervals of 0.03 seconds, especially those around 500 Hz. For a strongly distributed reference voice pattern and courtship emotions, a fundamental wave near 250 Hz appears, and harmonics up to around 6000 Hz appear, especially in a reference voice pattern and a threatening emotion with a strong distribution of components from 4000 to 6000 Hz. On the other hand, the component is distributed as a whole from 0 to 8000 Hz, and white noise-like sound lasts for about 0.8 seconds. Thus, a fundamental frequency is about 500 Hz, and a noise-like component is seen after about 0.2 seconds from the start of singing. Especially for a reference voice pattern having a strong tendency of a component around 4000 Hz, a feeling of joy, The basic frequency is 400 to 600 Hz, harmonics are distributed to around 3200 Hz, and a reference voice pattern having a characteristic that a slight vertical movement can be seen at the maximum fundamental frequency part. It is characterized by including at least one of the reference voice patterns that exist and particularly have a strong component distributed in the vicinity of 2000 to 4000 Hz.

  In addition, the cat emotion discrimination device includes a storage unit that stores a game program and game data, and an operation unit that is operated by a user, and the control unit is stored in the emotion-specific reference voice pattern storage unit. The success or failure of the user's judgment based on the cat's emotion determined from the comparison result between the reference voice pattern and the input voice pattern extracted by the input voice pattern extracting means, and the cat's emotion determined by the user by operating the above operating means In order to display the determination result on the display means, it is possible not only to determine the emotion of the cat but also to have game characteristics.

  According to the first aspect of the present invention, the cat's cry is converted into an electric voice signal, the characteristics of the relationship map between the time and frequency components of the voice signal are extracted as the input voice pattern, and the cat makes various emotions. Stores a reference voice pattern for each emotion that expresses the characteristics of the relationship map between the time and frequency components of the voice that is characteristically expressed as a voice, compares the input voice pattern with a reference voice pattern for each emotion, and compares them Thus, the emotion having the highest correlation with the input voice pattern is discriminated, so that the cat's emotion can be objectively discriminated from the emotion-specific reference voice pattern.

  According to the second aspect of the present invention, the reference sound pattern stored in the emotion-specific reference sound pattern storage means includes a storage means for storing the game program and the game data, and an operation means operated by the user. And the input voice pattern extracted by the input voice pattern extraction means, the cat's emotion determined from the result of the comparison, and the cat's emotion determined by the user by operating the operation means, to determine the success or failure of the user's judgment Since the determination result is displayed on the display means, not only the emotion of the cat is discriminated but also how much the user understands the emotion of the cat is displayed with game characteristics, Not only as a device for judging crying, it is possible to deepen the understanding of the cat from the crying of the cat.

  A cat emotion discriminator that compares the input cat voice with emotion-based reference voice patterns and displays the corresponding emotion discrimination results as images and voice, and converts the cat's cry into an electrical voice signal. Means, input voice pattern extracting means for extracting the characteristics of the relationship map between the time and frequency components of the voice signal as an input voice pattern, and the time and frequency components of the voice that the cat expresses various emotions characteristically as a cry A reference voice pattern storage means for storing emotions that express the characteristics of the relationship map for each of the various emotions, a comparison means for comparing the input voice patterns with the reference voice patterns for emotions, and the comparison Control means for determining the emotion most highly correlated with the input voice pattern and displaying the determination result on the image display means, and objectively assessing the cat's emotion from the cat's cry It was to be able to determine in.

  FIG. 1 shows a cat emotion discriminating apparatus (hereinafter referred to as an emotion discriminating apparatus) A based on a voice characteristic analysis of a cry according to the present invention, and this emotion discriminating apparatus A is formed as a whole imitating the form of a cat. The face portion is provided with a liquid crystal display 1 as image display means and a microphone 3 constituting a part of conversion means 2 for converting a cat's cry into an electrical sound signal. An operation means 5 operated by a user is arranged. Furthermore, in the body part, the input voice pattern extracting means 6 for extracting the characteristics of the relationship map between the time and frequency components of the voice signal as an input voice pattern, and the voice that the cat expresses various emotions characteristically as a cry A reference voice pattern storage unit by emotion for storing a reference voice pattern for each emotion expressing the characteristics of the relationship map of time and frequency components for each of the various emotions, and the input voice pattern is compared with the reference voice pattern for each emotion. A control means 10 having a comparison means 8 and an emotion discrimination means 9 for discriminating an emotion having the highest correlation with the input voice pattern is arranged, and the discrimination result discriminated by the emotion discrimination means 9 is displayed via the LCD driver 11 as an image. It is displayed on the liquid crystal display 1 which is a display means.

  The control means 10 may be composed of a control program for controlling the CPU and the entire emotion discriminating apparatus, a ROM that stores data such as character data, image data, and a reference voice pattern, and a work area RAM.

  The conversion means 2 converts the sound of a cat's sound into a digital sound signal, and is composed of a microphone 3 and an A / D converter 4. The cat's sound is received by the microphone 3 and converted into an electrical signal. Then, an audio signal obtained by digitizing the electric signal by the A / D converter 4 is generated.

  The operation means 5 erases all the set contents, the cursor key 5a used when selecting an arbitrary item from the selection items displayed on the screen, the cancel button 5b, and the decision button 5c for confirming the set contents. And a reset button 5d. The cancel button 5b can be turned ON / OFF by long pressing (for example, 2 seconds).

  The input voice pattern extraction means 6 extracts a characteristic pattern from the voice signal, and can be composed of a CPU (or a DSP), and stores a ROM storing a program for operating as the input voice pattern extraction means, What is necessary is just to provide RAM for work areas.

  By the way, an audio pattern is generally expressed in the form of a relationship map between time and frequency components of an audio signal, and this relationship map shows the temporal frequency distribution of audio, with the horizontal axis representing time and the vertical axis representing frequency. Preferably expressed in the form of a distribution of speech energy in individual grids divided by a fixed time interval and a fixed frequency interval. By expressing the relationship map in this way, comprehensive and quantitative handling of the audio signal becomes possible.

  The emotion-specific reference sound pattern storage means 7 stores reference sound patterns corresponding to various emotions, and may be composed of a ROM. This ROM may be a rewritable FLASH ROM, and data may be rewritten in response to future update of the reference voice pattern, addition of the number of emotions, and the like.

  The reference voice pattern is expressed in the form of a relationship map between the time and frequency components of the voice signal, similarly to the voice pattern. The relationship map is a representation of the temporal frequency distribution of speech expressed as time on the horizontal axis and frequency on the vertical axis, preferably individual grids divided by a constant time interval and a constant frequency interval. It is expressed in the form of the distribution of voice energy. Further, the reference voice pattern can be a pattern in which a highly universal characteristic part of the relation map is heavily weighted. By doing this, when comparing the reference voice pattern and the input voice pattern, even if it is a variety of input voice patterns, as long as it has a highly universal characteristic part corresponding to emotion, It becomes possible to make it correspond with the reference | standard audio | voice pattern corresponding to any emotion, and the certainty of discrimination | determination of an emotion can be improved.

  The comparison means 8 is a component that compares the input voice pattern with the emotion-specific reference voice pattern. The comparison means 8 may be constituted by a CPU (or a DSP), and is constituted by a ROM storing a program for operating the CPU as the comparison means 8, a work area RAM, and the like. The comparison can be performed by a technique of pattern matching the characterized pattern by a Hamming process. The result of the comparison is output as the level of correlation.

  The emotion discriminating means 9 discriminates an emotion corresponding to the reference voice pattern determined to have the highest correlation by comparing the input voice pattern and the emotion-specific reference voice pattern by the comparing means 8 as the cat's emotion. Is an element. The emotion discriminating means 9 may be constituted by a CPU (or a DSP), and is constituted by a ROM storing a program for operating the CPU as the emotion discriminating means 9, a work area RAM, and the like.

  In creating reference voice patterns corresponding to various emotions of cats, data on the utterances of cats when various emotions are collected for a large number of cats. Based on behavioral studies, it is sufficient to judge the behavior and attitude of the cat at that time. The data of many collected voices are classified according to emotions, and the voice patterns common to the voice data by those emotions are classified into the emotions. May be defined as a reference voice pattern corresponding to.

  In addition, this reference | standard audio | voice pattern can also be taken as what weighted big the characteristic part with high universality as mentioned above.

  In this embodiment, six types of emotions of “satisfaction”, “courting”, “intimidation”, “self-expression”, “joy”, and “request” are set as basic emotions of the cat. Explain the characteristics of the voice pattern for each emotion.

  For emotions of “satisfied”, as shown in FIG. 3, when the cat is satisfied with a voice pattern in which components are distributed in a pulsed manner at intervals of 0.03 seconds, particularly in the vicinity of 500 Hz. This is a sound pattern corresponding to the crying sound, and it often sounds like “gorologo”.

  As shown in FIG. 4, for the emotion of “courtship”, a cat has a voice pattern in which a harmonic wave up to around 6000 Hz appears with a fundamental wave around 250 Hz, and particularly the component from 4000 to 6000 Hz is strongly distributed. The sound pattern that corresponds to the cry that utters is often heard like “Gawah”.

  For the emotion of “intimidation”, as shown in FIG. 5, the cat is intimidated by an audio pattern in which components are distributed generally from 0 to 8000 Hz and white noise-like sound lasts for about 0.8 seconds. It often sounds like “sher” in an audio pattern that corresponds to the cry that is sometimes emitted.

For the emotion of “self-expression”, as shown in FIG. 6, the fundamental frequency is about 500 Hz, and a noise-like component is seen after about 0.2 seconds from the start of the sound, and particularly the component around 4000 Hz tends to be strong. A voice pattern corresponding to a voice uttered when a cat expresses itself is often heard as “Nyaon”.

  For the emotion of “joy”, as shown in FIG. 7, the fundamental frequency is 400 to 600 Hz, the harmonics are distributed up to around 3200 Hz, and a slight up and down movement is observed at the maximum fundamental frequency portion. The voice pattern corresponds to the cry that is produced when a cat expresses joy, and often sounds like "Funa".

  For the emotion of “request”, as shown in FIG. 8, the fundamental frequency is 400 to 600 Hz, the harmonics are distributed up to around 3200 Hz, and a slight up and down movement is seen at the maximum fundamental frequency portion. The voice pattern is a voice pattern that corresponds to the cry that the cat expresses a request, and often sounds like "Meow".

  Next, a usage mode of the emotion determination apparatus A having the above-described configuration will be described based on the flowcharts of FIGS.

  When the power is turned on, it is determined in step ST1 whether or not the initial setting has been made. If the initial setting has not been made, the process proceeds to step ST2 where the initial setting screen is displayed and the cat type, name, gender and personality are displayed. Use the cursor buttons to select from the screen that appears, and press the ENTER button to perform initial settings.

  When initial settings are made, an opening is displayed (step ST3), and then a menu screen is displayed (step ST4). Titles such as translation mode, gesture translation mode, health check mode, fortune-telling mode, partner mode are displayed. The mode is selected with the cursor button (step ST5), and the selected mode is executed when the enter button is pressed (step ST6).

  In the present embodiment, a translation mode and a partner mode for promoting the program based on the cat's emotion determined from the cat's cry from the above menu will be described.

  When the cat language translation is selected with the cursor button on the menu screen and the enter button is pressed, the process proceeds to the translation mode. After the title screen is displayed in step ST10, the process proceeds to step ST11 to enter a voice standby state. When the voice is received (step ST12), the process proceeds to step ST13 to perform voice analysis. In this voice analysis, the cat cry received by the conversion means 2 is converted into a digital electrical voice signal, and the input voice pattern extraction means 3 extracts a characteristic voice pattern from the converted voice signal. The extracted voice pattern is developed on the RAM in the form of a relation map.

  Next, the comparison means 8 reads out the reference voice patterns corresponding to the respective emotions stored in the emotion-specific reference voice pattern storage means 7, and compares it with the input voice patterns developed on the RAM. For example, a method of pattern matching using a hamming process can be used. By this comparison, the correlation between the input voice pattern and each emotion is quantified. Next, the emotion discriminating means 9 discriminates the emotion having the highest correlation value as the cat's emotion and proceeds to step ST14 to display the discriminated emotion on the display as a text (for example, in the case of a satisfactory emotion). After “Happy Nyan”), the process proceeds to step ST15 to repeatedly display the cat's facial expression (FIGS. 14A to 14C) expressing the emotion.

  The user can determine the cat's emotion from the cat's expression displayed on the display. To continue the translation mode, press the enter button in step ST16 to return to step ST11 and enter the voice standby state again. If the cancel button is pressed, the process proceeds to step ST17, along with an inquiry “Do you want to quit?” A selection message “Yes” or “No” is displayed on the display. Select “Yes” with the cursor button and press the Enter button to end the translation mode, return to the menu screen (step ST4), and change the mode again. Selection can be made, and if “NO” is selected and the OK button is pressed, the process returns to step ST11 to enter the voice standby state again.

  Select “Partner Mode” on the menu screen and press the Enter button to proceed to “Partner Mode”. After the title screen is displayed in Step ST20, the process proceeds to Step ST21 and becomes a selection screen. Use the cursor buttons to select the mode you want to execute from "Partner Degree" or "Recently Happy".

  If “I know you?” Is selected, the process proceeds to step ST101, enters the voice standby state, and receives the voice (step ST102). The process proceeds to step ST103, and the received voice is received in the same manner as in the translation mode described above. Perform analysis. When the analysis is completed, a question display screen is displayed (step ST104). From the displayed emotions, it is determined what emotion the current cat cry represents, and the user selects with the cursor button and determines with the decision button (step ST105). It is determined whether or not the emotion determined by the user and the emotion obtained by analyzing the voice match (step ST106). If they match, a correct animation is displayed in step ST107, and they match. If not, an incorrect animation is displayed in step ST108.

  If the partner mode is to be continued, pressing the enter button in step ST109 returns to step ST21 and the selection screen is displayed again. Therefore, the mode to be executed may be selected.

  When the cancel button is pressed, the process proceeds to step ST110, and a message “Yes” or “No” is displayed on the display together with an inquiry “Do you want to stop?” Select “Yes” with the cursor button and confirm If the button is pressed, the partner mode is ended, the menu screen (step ST4) is returned, and the mode can be selected again. If “NO” is selected and the enter button is pressed, the screen returns to the selection screen of step ST21. Partner mode can be run again.

  By repeating this routine, correct answers and incorrect answers are counted in the counter set in the memory, so that the correct answer rate can be determined.

  When confirming “partner level”, selecting “partner level” on the selection screen of step ST21 and pressing the enter button displays a result table (step ST201). In this result table, the number of problems (the number of times “I'm going to know?”), The number of correct answers, the number of incorrect answers, and the correct answer rate are displayed in a list (see FIG. 15A).

  After confirming, when the enter button is pressed, the process proceeds to step ST202, a message corresponding to the accuracy rate is read from the table, and the partner degree (for example, the name of the cat is registered as “tama” by default in the display) ), As shown in FIG. 15 (b), after “The compatibility between you and Tama-chan” is displayed, the summary results are displayed in comments (see FIG. 15 (c)). Is displayed (see FIG. 5D), when the confirmation determination button is pressed, the process returns to step ST109 to select whether to continue or end the partner mode.

  When confirming “Recent Dinner”, select “Recent Dinner” on the selection screen in Step ST21 and press the Enter button to display a screen as shown in FIGS. 16 (a) and 16 (d). Is displayed (step ST301). In this screen, the past 10 voice determinations are displayed as dots d in time series, and the more the dots d1 to d10 rise above the vertical axis, the better the mood, and the lower it goes down. You can see a list of bad moods, and you can recheck past voice judgments. In order to reconfirm, move the cursor c left and right to select an arbitrary dot d and press the enter button. Then, the process proceeds to steps ST302 and 303, where the comment and the cat's facial expression can be reconfirmed. ing. For example, as shown in FIG. 16 (a), when the cursor c is moved to the leftmost position, the dot d1 is selected. When the enter button is pressed, as shown in FIGS. 16 (b) and 16 (c), When the cursor c is moved to the right as shown in FIG. 16 (d), the dot d2 is selected as shown in FIG. 16 (d). As you can see, you can see a bad message and expression. When the confirmation button is pressed after confirmation, the process returns to step ST109 to select whether to continue or end the partner mode.

  As described above, when “translation mode” is selected and the cat's cry is detected, the cat's emotion is quickly analyzed and displayed on the screen, the partner mode is selected and the cat's cry is detected. At the same time, when the user decides what kind of emotion the call is, the user's judgment and the analysis result of the call determine how much the user understands the cat's emotion, The degree of understanding of cats is judged, so not only can you understand the feelings of the cat at that time but also the understanding of the user's cats, so that you can have a deeper partnership Tools can be provided.

The perspective view which shows an example of the emotion discrimination device of this invention The block diagram which shows the structure of the said emotion discrimination device It is a figure of the "time-frequency component relation map" showing the typical audio | voice pattern corresponding to the feeling of "satisfaction" (the horizontal scale is 0.2 second and the vertical scale is 500 Hz, (The relevant part is circled) It is a figure of the "time-frequency component relation map" showing the typical audio | voice pattern corresponding to the emotion of "courting" (one scale of a horizontal axis is 0.1 second, and one scale of a vertical axis is 500 Hz, (The relevant part is circled) It is a figure of the "time-frequency component relation map" showing a typical voice pattern corresponding to the emotion of "intimidation" (one scale on the horizontal axis is 0.1 second, one scale on the vertical axis is 500 Hz, (The relevant part is circled) It is a figure of the "time-frequency component relation map" showing the typical audio | voice pattern corresponding to the emotion of "self-expression" (the horizontal scale is 0.05 second, the vertical scale is 500 Hz, (Characteristic parts are circled) It is a figure of the "time-frequency component relation map" showing the typical audio | voice pattern corresponding to the emotion of "joy" (one scale of a horizontal axis is 0.1 second, and one scale of a vertical axis is 500 Hz, (The relevant part is circled) It is a figure of the "time-frequency component relationship map" showing the typical audio | voice pattern corresponding to the emotion of "request" (The horizontal scale is 0.2 second and the vertical scale is 500 Hz, (The relevant part is circled) The flowchart figure explaining the usage condition of an emotion discrimination device The flowchart figure explaining the usage condition of an emotion discrimination device The flowchart figure explaining the usage condition of an emotion discrimination device The flowchart figure explaining the usage condition of an emotion discrimination device The flowchart figure explaining the usage condition of an emotion discrimination device (A)-(c) are the display screens which show an example of the emotion of the cat displayed on a display (A) to (d) are display screens showing examples of screens displayed in the partner mode. (A)-(f) is a display screen showing an example of a screen displayed in the partner mode

Explanation of symbols

DESCRIPTION OF SYMBOLS 1 Image display means 2 Conversion means 5 Operation means 6 Input sound pattern extraction means 6
7 Reference voice pattern storage means by emotion 8 Comparison means 9 Emotion discrimination means 10 Control means

Claims (2)

A cat emotion discriminating apparatus that compares an input cat voice with an emotion-specific reference voice pattern and displays a corresponding emotion discrimination result as an image and voice,
Conversion means for converting the cat's cry into an electrical audio signal;
Input voice pattern extraction means for extracting the characteristics of the relationship map between the time and frequency components of the voice signal as an input voice pattern;
An emotion-specific reference voice pattern storage means for storing an emotion-specific reference voice pattern that represents the characteristics of the relationship map between the time and frequency components of the voice that the cat expresses various emotions characteristically as a cry;
A comparing means for comparing the input voice pattern with the reference voice pattern by emotion;
In the cat emotion discriminating apparatus based on the voice feature analysis of the cat's singing comprising the control means for judging the emotion most highly correlated with the input voice pattern by the comparison and displaying the judgment result on the image display means.
The above-mentioned standard voice patterns by emotion are roughly divided into satisfaction emotion, courtship emotion, threatening emotion, self-expression emotion, joy emotion, demand emotion,
For satisfied emotions, components are distributed in a pulse-like manner at intervals of 0.03 seconds, and in particular, a reference voice pattern in which components near 500 Hz are strongly distributed. For courtship emotions, the fundamental wave is around 250 Hz and is 6000 Hz. Harmonic waves up to the vicinity appear, especially in the reference voice pattern in which the component from 4000 to 6000 Hz is strongly distributed. For threatening emotions, the entire component from 0 to 8000 Hz is distributed, and the white noise-like sound is 0. Standard voice pattern that lasts for about 8 seconds For self-expressing emotions, the fundamental frequency is about 500 Hz, and a noise-like component is seen after about 0.2 seconds from the beginning of the sound, especially components around 4000 Hz tend to be strong The basic frequency is 400 to 600 Hz, and harmonics are distributed up to around 3200 Hz for pleasure emotions. A reference voice pattern having a characteristic that a slight vertical movement is seen at a portion where the fundamental frequency is maximum. For a required emotion, a reference voice pattern which exists near a fundamental frequency of 700 Hz and particularly has a strong component distributed around 2000 to 4000 Hz. A cat emotion discriminating apparatus based on a voice feature analysis of a cat cry characterized by including at least one of them. The cat emotion discriminating apparatus includes a storage unit storing a game program and game data, and an operation unit operated by a user, and the control unit stores a reference voice stored in the emotion-specific reference voice pattern storage unit. The pattern and the input voice pattern extracted by the input voice pattern extraction means are compared, and the user's judgment is determined based on the cat emotion determined from the comparison result and the cat emotion determined by the user by operating the operation means. The cat emotion discriminating apparatus according to claim 1, wherein success / failure is determined and a determination result is displayed on the display means.

Images