Speech analysis device, speech analysis system, speech analysis method, and speech analysis program

Abstract

To provide a speech analysis device, a speech analysis system, a speech analysis method, and a speech analysis program which can properly determine a user's personality, on the basis of the user's speech data.SOLUTION: A speech analysis device performs analysis based on a user's speech data. The speech analysis device comprises: a feature extraction unit which extracts a scale feature amount which is a feature amount for each scale in an octave sound range which is a sound range in units of an octave, by frequency analysis with respect to speech data; and a determination unit which associates a characteristic element specified based on determination reference which is reference based on feeling and action reference which is the reference based on a viewpoint with a scale, and determines the user's personality, on the basis of the scale feature amount in a consciousness octave sound range which is an octave sound range relevant to consciousness, a preconscious octave sound range which is the octave sound range relevant to preconscious, and a latent consciousness octave sound range which is the octave sound range relevant to latent consciousness.SELECTED DRAWING: Figure 1

Description

[Technical Field]

[0001] The present invention relates to a voice analysis device, a voice analysis system, a voice analysis method, and a voice analysis program that perform analysis based on a user's voice data, and in particular to a voice analysis device, a voice analysis system, a voice analysis method, and a voice analysis program that perform analysis related to the user's personality. [Background technology]

[0002] In recent years, emphasis has been placed on individuality in human relationships and work. Various attempts have been made to understand individuality, such as aptitude tests to recruit talented personnel and reduce turnover rates, and self-diagnosis to help people develop careers that suit them. If an individual is continually placed in relationships or a work environment that does not suit their personality, they may experience stress and strain, which can lead to poor health and depression. For companies, this can also lead to a decline in productivity due to a decline in employee motivation.

[0003] To understand individuality, questionnaire-based methods such as aptitude tests and stress checks are sometimes used. However, some aspects of individuality are not even consciously recognized by the individual, making it difficult to understand such aspects using methods such as questionnaires. Since individuality is inherently multifaceted, in order to build relationships and work environments suited to one's individuality, it is necessary to understand the individual's personality from multiple perspectives, including aspects of the individual's personality that the individual is not even consciously aware of. To understand individuality from multiple perspectives, some methods focus on the voice. This approach is based on the idea that voices come in a variety of types and characteristics and contain a great deal of information about the speaker's physical condition, emotions, thoughts, and so on, and claims that analyzing the voice can help understand individuality. Technologies have been proposed that attempt to understand individuality by analyzing such voices.

[0004] For example, Patent Document 1 proposes a voice analysis device that analyzes the voice emitted by an individual (speaker) to identify the characteristics of the speaker. The voice analysis device of Patent Document 1 identifies characteristic pitches based on the volume of each pitch included in the voice, and identifies the speaker's characteristics based on the identified pitches. In the embodiment described in Patent Document 1, the loudest pitch and the quietest pitch are used as characteristic pitches, and "color," "quality," "issue," and "song information" are used as feature information that classifies the speaker's characteristics, and feature information such as the speaker's "quality" is identified based on the identified pitches.

[0005] Patent Document 2 proposes a voice diagnostic device that analyzes and diagnoses the speaker's psychological state, health condition, or thought pattern based on the pitch of the speaker's voice. The voice diagnostic device of Patent Document 2 diagnoses the speaker's psychological state, health condition, or thought pattern based on the scale that maximizes and / or minimizes the proportion of the octave and the proportion of the scale related to the senses of the speaker, from voice frequency data converted from the speaker's voice data. The voice diagnostic device of Patent Document 2 makes a diagnosis using the relationship between the pitch of the voice produced by a human and the five human senses: taste, touch, smell, hearing, and sight.

[0006] Patent Document 3 proposes a voice diagnostic system that applies the voice analysis device of Patent Document 1 to voices emitted by multiple individuals (speakers). The processing content of the voice diagnostic system of Patent Document 3 is substantially the same as that of the voice analysis device of Patent Document 1. The voice diagnostic system of Patent Document 3 uses the relationship between the pitch of a human voice and human senses (five senses) in the same way as the voice diagnostic device of Patent Document 2 to diagnose the speaker's psychological state, health state, or thought pattern, but does not disclose a specific diagnostic method. [Prior art documents] [Patent documents]

[0007] [Patent Document 1] Japanese Patent Application Laid-Open No. 2007-212932 [Patent Document 2] Patent No. 6029223 [Patent Document 3] Japanese Patent Publication No. 2022-145373 Summary of the Invention [Problem to be solved by the invention]

[0008] However, the voice analysis device of Patent Document 1 identifies a characteristic pitch and regards the characteristic information corresponding to the identified pitch as the speaker's characteristics, and since the speaker's characteristics are grasped only from the pitch and characteristic information, there is a possibility that the speaker's multifaceted personality cannot be fully grasped. The voice diagnosis device of Patent Document 2 diagnoses the speaker's psychological state, health state, or thought pattern using the relationship between musical scales and human senses, and since the diagnosis is made only from musical scales and senses, there is also a possibility that the speaker's personality cannot be fully grasped. The voice diagnosis system of Patent Document 3 performs the same processing as the voice analysis device of Patent Document 1, and therefore has the same possibilities as the voice analysis device of Patent Document 1.

[0009] The present invention has been made in light of the above-mentioned circumstances, and an object of the present invention is to provide a voice analysis device, a voice analysis system, a voice analysis method, and a voice analysis program that can accurately determine a user's personality based on the user's voice data. [Means for solving the problem]

[0010] The present inventors have conducted extensive research to solve the above problems, and as a result have found that the following invention meets the above object, thereby completing the present invention. That is, the present invention relates to the following inventions.

[0011] <1> A voice analysis device that performs analysis based on voice data of a user, comprising: a feature extraction unit that extracts scale features, which are features for each scale in an octave range, which is a range of sound in octaves, by frequency analysis of the voice data; and a judgment unit that associates characteristic elements and scales defined based on judgment criteria, which are criteria based on sensations, and behavioral criteria, which are criteria based on viewpoints, and judges the personality of the user based on the scale features in the conscious octave range, which is an octave range related to the conscious mind, the preconscious octave range, which is an octave range related to the preconscious mind, and the subconscious octave range, which is an octave range related to the subconscious mind. <2> The determination unit determines the individuality of the user by determining the scale feature extracted by the feature extraction unit using a threshold value for the scale feature. <1> The voice analysis device according to claim 1. <3> The determination unit determines the type of the user by defining a plurality of types based on the determination criteria and the behavioral criteria and determining the scale feature of the subconscious octave range extracted by the feature extraction unit using the threshold value. <2> The voice analysis device according to claim 1. <4> The determination unit defines an ability exertion level based on information regarding a difference between the scale feature amount of the conscious octave range and the scale feature amount of the subconscious octave range, and determines the ability exertion level of the user based on the scale feature amount of the conscious octave range and the scale feature amount of the subconscious octave range extracted by the feature extraction unit. <1> from <3> 10. The speech analysis device according to claim 9, wherein <5> The apparatus further includes a display unit that displays the result of the determination by the determination unit. <1> from <4> 10. The speech analysis device according to claim 9, wherein <6> The display unit displays the scale feature extracted by the feature extraction unit for each of the characteristic elements. <5> The voice analysis device according to claim 1. <7> The display unit displays the scale feature quantity using a color defined for each scale. <6> The voice analysis device according to claim 1. <8> The voice data is data in which the user speaks his / her own name. <1> from <7> 10. The speech analysis device according to claim 9, wherein <9> <1> from <4> and a terminal device used by the user, wherein the terminal device transmits the user's voice data to the voice analysis device, and the voice analysis device transmits the result of the judgment made by the judgment unit to the terminal device. <10> A voice analysis method for performing an analysis based on voice data of a user, the voice analysis method comprising: a feature extraction step of extracting, by frequency analysis of the voice data, scale features which are features for each musical scale in an octave range, which is a range of sound in octaves; and a judgment step of associating characteristic elements defined based on judgment criteria, which are criteria based on sensations, and behavioral criteria, which are criteria based on viewpoints, with musical scales, and judging the personality of the user based on the scale features in the conscious octave range, which is an octave range related to the conscious mind, the preconscious octave range, which is an octave range related to the preconscious mind, and the subconscious octave range, which is an octave range related to the subconscious mind. <11> The voice data is data in which the user speaks his / her own name. <10> The speech analysis method according to claim 1. <12> In the determining step, the personality of the user is determined for the purpose of counseling the user. <10> or <11> The speech analysis method according to claim 1. <13> In the determining step, the personality of the user is determined for counseling including advice on health and beauty for the user. <12> The speech analysis method according to claim 1. <14> A degree of performance of the user is calculated based on information about a difference between the scale feature amount of the conscious octave range and the scale feature amount of the subconscious octave range. <10> from <13> 2. A speech analysis method according to claim 1, wherein <15> On the computer, <10> from <14> 10. A speech analysis program for executing any one of the speech analysis methods described above. [Effects of the Invention]

[0012] According to the voice analysis device, voice analysis system, voice analysis method, and voice analysis program of the present invention, the user's personality is judged using characteristic elements defined based on judgment criteria and behavioral criteria, based on information such as the characteristics of each scale and the octave range related to consciousness, thereby making it possible to accurately judge a multifaceted personality. [Brief explanation of the drawings]

[0013] [Figure 1] 1 is a block diagram showing an example of the configuration (first embodiment) of a voice analysis device according to the present invention. [Figure 2] FIG. 10 is an image diagram showing an example of a scale information table stored in a storage unit. [Figure 3] FIG. 10 is a diagram illustrating an example of a type classification table. [Figure 4] FIG. 10 is a conceptual diagram illustrating an example of a type information table stored in a storage unit. [Figure 5] FIG. 2 is a block diagram showing an example of the configuration of a determination unit (first embodiment). [Figure 6] FIG. 10 is a diagram showing an example of a graph displaying scale features. [Figure 7] 1A and 1B are diagrams showing examples of radar charts for displaying scale features, where (A) is a diagram before the scale features are displayed, and (B) is a diagram after the scale features are displayed. [Figure 8] 10A and 10B are diagrams showing examples of displaying type characteristics, in which (A) is a diagram showing a display field for displaying type characteristics, and (B) is a diagram showing an example of displaying type D. FIG. [Figure 9] 4 is a flowchart showing an operation example (first embodiment) of the voice analysis device. [Figure 10] FIG. 1 is a block diagram showing an example of the configuration of a voice analysis device according to the present invention (second embodiment). [Figure 11] FIG. 10 is a block diagram showing an example of the configuration of a determination unit (second embodiment). [Figure 12] FIG. 10 is a diagram showing an example of the configuration of a voice analysis system according to the present invention (third embodiment). [Figure 13] FIG. 10 is a block diagram showing an example of the configuration of a voice analysis device (third embodiment). DETAILED DESCRIPTION OF THE INVENTION

[0014] In this invention, a user's personality is determined based on the user's voice data. Personality is defined as the unique characteristics of an individual or an individual thing, and indicates various characteristics that a person possesses, such as their strengths and weaknesses, aptitudes, strengths and weaknesses, etc. In this invention, a user's personality is determined based on voice data in which the user speaks their own name, etc.

[0015] In this invention, frequency analysis is performed on the user's voice data to extract features (scale features) for each scale in a range measured in octaves (octave range). An octave is defined as the eighth note from a given note in the diatonic scale, or the distance between those notes, where the ratio of the vibration frequencies of two notes (a given note and the eighth note) is 1:2. Specifically, "a note with twice the frequency of a given note" corresponds to "a note one octave higher," and "a note with half the frequency of a given note" corresponds to "a note one octave lower." In the present invention, an "octave range" refers to a range of notes separated by this interval, and the octave range to which a certain note belongs does not include the notes one octave above or one octave below that note; the notes belong to the octave ranges above and below that note, respectively. Generally, in Western music, musical scales are based on octaves, and one octave is made up of 12 notes (seven whole tones and five semitones). In the present invention, for example, in this configuration, the features of each scale are extracted as scale features through frequency analysis. Note that there are various types of octave ranges and scales, and the present invention can accommodate various octave ranges and scales.

[0016] In the present invention, the user's personality is judged based on the scale feature amounts in the octave range related to human consciousness. Specifically, the user's personality is judged based on the scale feature amounts in the octave range related to the conscious mind, which is the consciousness of which the user is aware (conscious consciousness octave range), the octave range related to the subconscious mind, which is the consciousness of which the user is not aware (subconscious consciousness octave range), and the octave range related to the preconscious mind, which is the interference area between the conscious mind and the subconscious mind (preconscious consciousness octave range).

[0017] Furthermore, in this invention, elements (characteristic elements) that constitute individuality, which are defined based on judgment criteria and behavioral criteria, are associated with each musical scale and used to judge a user's individuality. The judgment criteria are based on sensations and are used to identify which sensation is dominant in the user and where the judgment is made, i.e., the dominant sensation. The behavioral criteria are based on perspectives and are used to identify for whom the user acts and who has a perspective, i.e., the starting point of the action and the target of the perspective.

[0018] In this way, the present invention judges a user's personality based on musical scales and information on consciousness, judgment criteria, and behavioral criteria, making it possible to accurately and broadly judge the user's personality. The judgment results can be used for counseling, such as advice on mental health, health, and beauty.

[0019] In the present invention, a threshold is set for the scale feature, and a user's personality can be determined by threshold judgment of the scale feature extracted from the user's voice data. Furthermore, a plurality of types can be defined based on judgment criteria and behavioral criteria, and a user's type can be determined by threshold judgment of the scale feature of the subconscious octave range. Furthermore, in the present invention, a performance level indicating how well the user is demonstrating his or her abilities can be calculated based on information regarding the difference between the scale feature of the conscious octave range and the scale feature of the subconscious octave range.

[0020] In the present invention, it is possible to display the judgment results such as the scale feature amount for each characteristic element and the user type, etc. By displaying the scale feature amount in color and the judgment results in a graph, etc., the results can be communicated to the user in an easy-to-understand manner.

[0021] The present invention may be realized as a speech analysis device that performs the above-described processing, or as a speech analysis system that includes a speech analysis device and a terminal device that transmits user speech data. In this case, it is also possible to display the results of the speech analysis device's judgment on the terminal device. This allows the user to have their personality judged from any location, such as a location remote from the speech analysis device. The present invention may be realized as a speech analysis method that performs the above-described processing, or as a speech analysis program that causes a computer to execute the speech analysis method.

[0022] Hereinafter, embodiments of the present invention will be described with reference to the drawings. In each drawing, the same components are designated by the same reference numerals, and their description may be omitted. Furthermore, the configurations and the like in the following description are examples, and the present invention is not limited to these. Furthermore, the following description focuses on the main configurations and operations for implementing the present invention, and descriptions of general-purpose processing required for implementing the present invention, such as communication processing, may be simplified or omitted.

[0023] Fig. 1 shows an example of the configuration of a voice analysis device according to the present invention (first embodiment). The voice analysis device 1 of this embodiment uses the voice (voice data) spoken by the user to determine the user's personality and displays the determination result. In determining the personality, the type to which the user belongs is determined from a plurality of predefined types, and health and beauty advice according to the type is output, which can be used for counseling the user.

[0024] The voice analysis device 1 defines individuality from the perspectives of "sensation," "perspective," and "awareness," and determines the individuality of the user.

[0025] From the perspective of "sensation," personality is defined based on which sense is dominant among human senses, where decisions are made, where information is obtained, etc. The voice analysis device 1 uses the following senses: bodily sensations (touch, smell, taste), auditory sensations (hearing), and visual sensations (sight, visual intuition, intuition).

[0026] From the perspective of "perspective," personality is defined based on the target person who is the starting point of an action or the viewpoint, such as for whom one acts, who has a viewpoint, and for whom one is working hard. The voice analysis device 1 classifies targets into three categories: oneself, others, and society, and defines personality as a self-axis, others-axis, and society-axis, respectively. The self-axis indicates that one has a viewpoint and tends to act for oneself, the others-axis indicates that others have a viewpoint and tends to act for others, and the society-axis indicates that everyone has a viewpoint and tends to act for everyone.

[0027] From the perspective of "consciousness," individuality is defined based on human consciousness, such as in which conscious area one feels and where consciousness is located. The voice analysis device 1 divides consciousness into three areas: conscious, subconscious, and preconscious. The conscious mind is the consciousness that a person is aware of, and is the thought area corresponding to the current state of thought. The subconscious mind is the consciousness that a person is not aware of, and is the quality / essence area corresponding to the person's innate qualities and essence. The preconscious mind is the part of consciousness where the conscious mind and subconscious mind interfere, and is the part where things that are done consciously can be done unconsciously, and is the habit / habit area that mainly manifests as habits and quirks.

[0028] The voice analysis device 1 defines characteristic elements that make up personality for each of the three consciousnesses (conscious, preconscious, and subconscious) based on the judgment criteria and the behavioral criteria, with criteria based on sensations as the judgment criteria and criteria based on viewpoints as the behavioral criteria. Furthermore, personality is classified into multiple types based on the judgment criteria and the behavioral criteria, and the characteristics of each type are defined.

[0029] The voice analysis device 1 includes a feature extraction unit 11 that extracts scale features in an octave range by frequency analysis of the user's voice data, a judgment unit 12 that judges the user's personality based on scale features in a conscious octave range related to the conscious mind, a preconscious octave range related to the preconscious mind, and a subconscious octave range related to the subconscious mind, a display unit 13 that displays the results of the judgment by the judgment unit 12, and a memory unit 14 that stores various information used by the feature extraction unit 11 and the judgment unit 12 as a table.

[0030] The feature extraction unit 11 inputs the user's voice data Vd and gender information Gi indicating the user's gender, and outputs thought octave features Tof including scale features in the conscious octave range, habit octave features Hof including scale features in the preconscious octave range, and quality octave features Nof including scale features in the subconscious octave range.

[0031] For example, a user speaks their own name for a predetermined period of time (e.g., six seconds). Since a name defines oneself, it is preferable to use data of the user's spoken name as the voice data used to determine personality. The voice spoken by the user is input as voice data Vd to the feature extraction unit 11 via a voice input means such as a microphone provided in the voice analysis device 1. Furthermore, gender information Gi indicating the user's gender (male or female) is input to the feature extraction unit 11 via a data input means such as a keyboard provided in the voice analysis device 1. The user's voice may be recorded using a device other than the voice analysis device 1, and the recorded voice may be input to the feature extraction unit 11 as voice data Vd. Furthermore, the user's gender, etc. may be registered in advance in the voice analysis device 1, and when determining the user's personality, the registered gender may be input to the feature extraction unit 11 as gender information Gi.

[0032] The feature extraction unit 11 applies a Fourier transform, which is a frequency analysis technique, to the input voice data Vd to convert the voice data Vd, which is time domain data, into frequency domain data (hereinafter referred to as "amplitude data"). The Fourier transform may be a discrete Fourier transform or a fast Fourier transform, which are commonly used for digital signals. Note that since silent intervals (intervals where no voice is uttered) may exist before and after the voice data Vd, these silent intervals may be deleted when converting the voice data Vd into amplitude data. Furthermore, a frequency analysis technique other than the Fourier transform may be used to convert the voice data Vd into amplitude data.

[0033] The feature extraction unit 11 divides the frequency domain in which the amplitude data is distributed into eight octave ranges, assuming that the voice uttered by the user has an eight-octave range, and further divides each octave range into 12 scale ranges (hereinafter referred to as "scale ranges"), assuming that one octave is made up of 12 scales. Hereinafter, the eight octave ranges will be expressed as octave 0, octave 1, ..., octave 7 in order of decreasing frequency, and the 12 scales will be expressed as C, C#, D, D#, E, F, F#, G, G#, A, A#, B. The scales in each octave range will be expressed by adding the last number of the octave range notation to the end of the scale notation, for example, the scale C in octave 0 will be expressed as C0, and the scale A# in octave 7 will be expressed as A#7. When there is no need to distinguish between octave ranges, the scale is generally written as C to B.

[0034] A specific division method involves determining the center value of each scale in each octave range (hereinafter referred to as the "scale center frequency"), and setting the frequency midway between the scale center frequencies of two adjacent scales as the boundary value of the scale range of the two scales. For example, if the scale center frequencies of C1, C#1, and D1 are approximately 65 Hz, 69 Hz, and 73 Hz, respectively, the boundary values ​​are approximately 67 Hz and 71 Hz, and the scale range of C#1 is approximately 67 to 71 Hz. For example, in octave 0, the feature extraction unit 11 sets the scale center frequency of C0 at approximately 33 Hz, and values ​​incrementing that value by 2 to 3 Hz are set as the scale center frequencies of the subsequent C#0 to B0 notes. For octaves 1 to 7, the scale center frequency of each scale in each octave range is set as twice the scale center frequency of the same scale in the previous octave range. When set in this way, the scale center frequency of C7 is approximately 4186 Hz, and the scale center frequency of B7 is approximately 7902 Hz. Such setting of the scale center frequencies is done based on tuning such as equal temperament.

[0035] The feature extraction unit 11 divides each scale range of each octave range into a predetermined number (e.g., 350) and calculates the average value (hereinafter referred to as "channel amplitude") of amplitude data included in the divided ranges (hereinafter referred to as "channel ranges"). In the case of eight octave ranges, twelve scale ranges, and 350 channel ranges, the feature extraction unit 11 calculates 33,600 channel amplitudes from the amplitude data. The feature extraction unit 11 then calculates the ratio (percentage) of each channel amplitude to the maximum channel amplitude in the octave range including the channel range corresponding to that channel amplitude (hereinafter referred to as "channel amplitude ratio"). For example, if the maximum channel amplitude in octave 1 is 50 and the channel amplitude in one channel range at C1 is 15, the channel amplitude ratio in that channel range is 30 (= 15 / 50 × 100). In order to suppress the high-frequency channel amplitude ratios of the obtained channel amplitude ratios, the feature extraction unit 11 may perform high-pass filtering to gradually decrease the magnitude of the channel amplitude ratios in the high-frequency scale range (e.g., E5 to B7). For example, the feature extraction unit 11 performs high-pass filtering by multiplying the channel amplitude ratios in the high-frequency scale range by a multiplier that gradually decreases from 1 to 0.4. In cases where it is not necessary to suppress the high-frequency channel amplitude ratios, the feature extraction unit 11 does not need to perform high-pass filtering.

[0036] The feature extraction unit 11 calculates, for each scale range in each octave range, a value obtained by integrating all channel amplitude ratios included in that scale range (hereinafter referred to as "scale integral value"). In the case of eight octave ranges, 12 scale ranges, and 350 channel ranges, 96 scale integral values ​​are calculated.

[0037] As described above, the voice analysis device 1 determines personality based on characteristic elements defined in each of the three consciousnesses (conscious, preconscious, and subconscious). In determining personality, the octave ranges associated with each of the three consciousnesses (conscious octave range, preconscious octave range, and subconscious octave range (hereinafter sometimes collectively referred to as "conscious octave range")) are determined in advance based on knowledge. Multiple octave ranges may be associated with one consciousness, and all of these may be designated as the conscious octave range, or just one of them may be designated as the conscious octave range. Furthermore, based on the idea that low-frequency octave ranges are associated with the subconscious and high-frequency octave ranges are associated with the conscious mind, the octave ranges associated with each consciousness may differ depending on gender. For example, in the voice analysis device 1, for males, octaves 6 and 7 are associated with the conscious mind, octaves 2 and 3 are associated with the preconscious mind, and octaves 0 and 1 are associated with the subconscious mind; for females, octaves 6 and 7 are associated with the conscious mind, octaves 3 and 4 are associated with the preconscious mind, and octaves 1 and 2 are associated with the subconscious mind; for males, octave 7 is the conscious octave range, octave 3 is the preconscious octave range, and octave 0 is the latent octave range; for females, octave 7 is the conscious octave range, octave 4 is the preconscious octave range, and octave 1 is the latent octave range. The correspondence between consciousness and octave range is stored in advance in the storage unit 14 as a consciousness correspondence information table.

[0038] The feature extraction unit 11 determines the gender of the user from the input gender information Gi, and reads out information on the octave range, which is the conscious octave range, preconscious octave range, and latent octave range for the determined gender, as consciousness octave information Coi from the consciousness correspondence information table in the storage unit 14. For example, when the gender information Gi is "female," information that the conscious octave range, preconscious octave range, and latent octave range are octaves 7, 4, and 1, respectively, is read out as consciousness octave information Coi.

[0039] The feature extraction unit 11 confirms the octave range corresponding to each conscious octave range using the conscious octave information Coi, and then performs normalization processing (normalization) so that the scale integral value in each conscious octave range becomes an integer value within a predetermined range (e.g., an integer value between 0 and 10). For example, an adjustment value by which the scale integral value in each conscious octave range is divided is determined in advance for each gender, and the scale integral value in each conscious octave range is divided by the corresponding adjustment value, and the division result is converted into an integer. The adjustment value is calculated in advance using previously collected voice data, etc., and stored in the consciousness correspondence information table stored in the storage unit 14 along with the correspondence between consciousness and octave range. For example, 28, 27, and 24 are stored for men, and 21, 32, and 25 are stored for women as adjustment values ​​for the conscious octave range, preconscious octave range, and latent octave range, respectively.

[0040] The integer conversion is performed by, for example, setting the division result to 0 if it is less than 1, and rounding off the decimal point if it is 1 or more. If the division result exceeds 10, it is rounded up to 10. The scale integral value after normalization becomes the scale feature. Note that the adjustment value may be changeable to deal with cases where the division result frequently exceeds 10.

[0041] As in the case of inputting the conscious octave information Coi, the feature extraction unit 11 determines the user's gender from the gender information Gi and reads out the adjustment values ​​for each conscious octave range for the determined gender as adjustment value information Nvi from the consciousness correspondence information table in the storage unit 14. For example, if the gender information Gi is "male," adjustment values ​​28, 27, and 24 for the conscious octave range, preconscious octave range, and latent octave range, respectively, are read out as adjustment value information Nvi. Using the read-out adjustment values, the feature extraction unit 11 performs a normalization process on the scale integral values ​​for each conscious octave range to obtain scale features. The scale feature for the conscious octave range is then output as a thinking octave feature Tof, the scale feature for the preconscious octave range as a habit octave feature Hof, and the scale feature for the subconscious octave range as a quality octave feature Nof. The feature extraction unit 11 may read out the adjustment value information Nvi at the same time as reading out the intentional octave information Coi.

[0042] The judgment unit 12 receives the thinking octave feature Tof, the habit octave feature Hof, and the quality octave feature Nof, judges the personality of the user, and outputs the judgment result as personality information Pi and type information Ti.

[0043] As described above, the voice analysis device 1 judges personality based on characteristic elements defined in each of the three consciousnesses (conscious, preconscious, and subconscious) on the basis of judgment criteria based on sensations and behavioral criteria based on viewpoints. The characteristic elements are associated with musical scales, and each musical scale is also associated with a color that represents its characteristics. The correspondence between musical scales, colors, and characteristic elements is stored in advance in the storage unit 14 as a musical scale information table.

[0044] An example of a scale information table is shown in Figure 2. In Figure 2, the color corresponding to each scale is listed in the second column, and the characteristic elements corresponding to each scale in the subconscious, preconscious, and conscious mind, respectively, are listed in the third, fourth, and fifth columns.

[0045] Regarding the relationship between musical scales and colors, a phenomenon called color hearing, which is a type of synesthesia, is known and various studies have been conducted. Based on this knowledge, the voice analysis device 1 defines the colors corresponding to each musical scale as shown in Figure 2. The colors corresponding to each musical scale are also related to the characteristic elements corresponding to that musical scale, and the voice analysis device 1 judges personality by closely associating the musical scale, color, and characteristic elements.

[0046] The characteristic elements are defined based on criteria and behavioral standards. The judgment criteria are based on sensations, and the voice analysis device 1 uses the somatic sensations (touch, smell, taste), auditory sensation (hearing), and visual sensations (sight, visual intuition, intuition) as the sensations. In terms of the correspondence between these three sensations and musical scales (colors), it is considered that the somatic sensations are related to C (red) to D# (gold), the auditory sensations are related to E (yellow) to G (turquoise blue), and the visual sensations are related to G# (blue) to B (magenta pink), and this idea is reflected in the definition of the characteristic elements. The behavioral standard is a standard based on a viewpoint, and the voice analysis device 1 uses the self-axis, the other-axis, and the social axis as viewpoints. In the correspondence between these three axes and musical scales (colors), the self-axis is related to E (yellow), the other-axis is related to G (turquoise blue), and the social axis is related to B (magenta pink), and this idea is reflected in the definition of the characteristic elements.

[0047] The characteristic elements are defined in each of the conscious, preconscious, and subconscious minds, and are basically defined based on the ideas of the judgment criteria and behavioral standards mentioned above, but some of them are expressed differently to suit the characteristics of each consciousness, as shown in Figure 2. The colors and characteristic elements shown in FIG. 2 are merely examples, and other definitions are also possible.

[0048] Furthermore, as described above, the voice analysis device 1 classifies personalities into multiple types based on the judgment criteria and behavioral criteria, and determines the type of the user. The type classification is performed by combining the judgment criteria with at least the self-axis, other-axis, and social axis used in the behavioral criteria. The voice analysis device 1 classifies personalities into 13 types, for example, as shown in FIG. 3 (hereinafter, the table shown in FIG. 3 is referred to as the "type classification table"). In the type classification table shown in FIG. 3, the behavioral criteria and judgment criteria elements related to each type are listed for each of the 13 types (Type A to Type M), and a user whose type has been determined has the elements listed in the determined type. "-" indicates that there is no specific related element. Note that the types shown in FIG. 3 are just an example, and other type classifications are also possible.

[0049] The voice analysis device 1 defines the characteristics of each type, and the details are stored in advance as a type information table in the storage unit 14. The voice analysis device 1 defines the characteristics of each type by dividing them into items such as "type name," "characteristics," "strengths," "weaknesses," "fatigue type name," "fatigue type description," and "health and beauty advice," and the type information table describes the specific details of each item for each type.

[0050] Figure 4 shows an example of the characteristics of each of types A to E as a representative example of a type information table. In Figure 4, the specific details of the characteristics for each type are listed in the following order: "Type name," "Characteristics," "Strengths," "Weaknesses," "Fatigue type name," "Explanation of fatigue type," and "Health and beauty advice." From the details listed in the type information table, users can understand the specific details of the characteristics of their own type.

[0051] The judgment unit 12 performs a process of judging the personality of a user based on characteristic elements (hereinafter referred to as "characteristic analysis process") and a process of judging the type of the user (hereinafter referred to as "type judgment process"). Fig. 5 shows an example of the configuration of the judgment unit 12, which includes a characteristic analysis unit 121 that performs characteristic analysis process and a type judgment unit 122 that performs type judgment process. The thinking octave feature Tof and habit octave feature Hof input to the judgment unit 12 are input to the characteristic analysis unit 121, and the quality octave feature Nof is input to the characteristic analysis unit 121 and the type judgment unit 122.

[0052] The characteristic analysis unit 121 reads out the data (color, characteristic elements) in the scale information table stored in the storage unit 14 as scale information Si1, associates the data of each scale in the conscious mind with the scale feature amount contained in the thinking octave features Tof, associates the data of each scale in the preconscious mind with the scale feature amount contained in the habit octave features Hof, and associates the data of each scale in the subconscious mind with the scale feature amount contained in the quality octave features Nof, and outputs the result as individuality information Pi. The magnitude of the scale feature amount indicates the strength of the characteristic element associated with that scale feature amount, and the larger the scale feature amount, the stronger the user's possession of the corresponding characteristic element.

[0053] The type determination unit 122 performs type determination processing using the quality octave feature Nof. Based on the idea that the subconscious is the domain of the inherent quality essence that a person possesses and is an important domain that determines characteristics and aptitude, the type of the user is determined using the quality octave feature Nof, which includes the scale feature amount in the subconscious octave range.

[0054] To determine the user's type, the type determination unit 122 preliminarily sets a threshold value for the scale feature (e.g., a threshold value of 2) and performs a determination using the threshold value on the scale feature included in the quality octave feature Nof. The type determination unit 122 reads out, from the scale information table stored in the storage unit 14, characteristic elements of the scale corresponding to the scale feature values ​​that exceed the threshold value, as scale information Si2. Then, based on the read scale information Si2, the type determination unit 122 determines the user's type in accordance with the type classification table shown in FIG. 3. For example, if only the scale feature values ​​of the G and G# scales exceed the threshold value, the user is determined to have the partner axis and vision / thinking, which are characteristic elements of the G and G# scales, and the user's type is determined to be Type D. The type determination unit 122 reads out the characteristics of the determined type from the type information table stored in the storage unit 14 as type information Ti and outputs it.

[0055] The display unit 13 receives the individuality information Pi and the type information Ti, and displays them on a display means such as a display provided in the voice analysis device 1 as the result of the judgment of the user's individuality.

[0056] The display unit 13 displays the scale feature as a graph for each of the three consciousnesses (subconscious, preconscious, and conscious) based on the personality information Pi, as shown in FIG. 6, for example. In FIG. 6, bar graphs showing the scale feature, with the scale feature on the vertical axis and the characteristic element on the horizontal axis, are displayed side by side for each consciousness. By displaying the scale feature as a graph, it is possible to easily grasp the characteristic elements prominent in each consciousness and to easily grasp the differences between the consciousnesses. In the bar graph shown in FIG. 6, the bar for each characteristic element may be displayed using a color corresponding to that characteristic element. Using colors also makes it possible to express an overall image. Furthermore, the threshold value used by the type determination unit 122 may be included in the personality information Pi, and in addition to the bar graph for the subconscious, a line may be drawn at the threshold position to indicate the threshold value in the bar graphs for the preconscious and conscious consciousness. This makes it easy to grasp the characteristic elements whose scale feature value exceeds the threshold value, helping to understand the user's personality.

[0057] The display unit 13 may display the musical scale feature quantities in a radar chart as shown in Fig. 7(A). The radar chart shown in Fig. 7(A) is in the shape of a regular dodecagon, with each vertex corresponding to a musical scale. The center of the regular dodecagon is connected to each vertex by a line segment, and scales are set at equal intervals from the center to each vertex. The musical scale feature quantities are plotted for each consciousness on this radar chart and connected by lines to display the data of the personality information Pi.

[0058] Figure 7(B) shows an example of a radar chart display corresponding to the example graph displaying the scale features shown in Figure 6. In Figure 7(B), the scale features of the subconscious (subconscious octave range) are displayed by a solid line L1, the scale features of the preconscious (preconscious octave range) by a dashed line L2, and the scale features of the conscious (conscious octave range) by a dashed line L3. In Figure 7(B), the scale features of each consciousness are distinguished by different line types, but they can also be distinguished by different line colors. Furthermore, if possible, the inside of the lines can be colored semitransparently so that they can be seen even when overlapping, and the data of the personality information Pi can be displayed. Using such a radar chart makes it easy to visualize the personality characteristics of each consciousness of the user, and clearly expresses the differences between consciousnesses.

[0059] The display unit 13 displays the user's personality type and its characteristics based on the type information Ti. For example, as shown in FIG. 8(A), a display column is provided for each type characteristic item, and the type characteristics included in the type information Ti are displayed. In FIG. 8(A), the "Type" column displays the type name in the upper row and the type's characteristics in the lower row, the "Strengths" column displays strengths, the "Weaknesses" column displays weaknesses, the "Fatigue Type" column displays the fatigue type name in the upper row and an explanation of the fatigue type in the lower row, and the "Health and Beauty" column displays health and beauty advice. As an example, FIG. 8(B) shows a display example when the user is determined to be Type D. For Type D, the "Type" column will display "Intelligent type" and "A person with an objective perspective and excellent thinking ability. This type is intelligent and is constantly taking in and processing a lot of information." The "Strengths" column will display "An intelligent person with an objective perspective who can think calmly and deeply. They are also good at visual tasks such as IT and design. They are also able to see things from the other person's perspective and work hard for them." The "Weaknesses" column will display "Because they have high thinking ability, they also tend to overthink. They are also reserved and sometimes find it difficult to take leadership roles or communicate through conversation." The "Fatigue type" column will display "Information fatigue type" and "Because they are always catching and processing a lot of information, they tend to have little time to rest both physically and mentally. This type is prone to fatigue due to information overload." The "Health and beauty" column will display "Releasing stiffness around the eyes and head, which are often overworked, will also help relieve fatigue throughout the body. Also, try to make a conscious effort to take time away from your smartphone or computer."

[0060] In this way, the display unit 13 displays the judgment results of the user's personality in an easy-to-understand manner using graphs, tables, etc., allowing the user to accurately grasp their own personality. Furthermore, when providing counseling to the user, the display is easy for the user to understand, making it possible to provide highly convincing counseling, and also to provide fatigue type and health and beauty advice, making it possible to provide a wide range of counseling.

[0061] The memory unit 14 stores the above-mentioned consciousness correspondence information table, scale information table, and type information table. The memory unit 14 may also be used as a temporary storage area for data used and / or calculated in the processes of the feature extraction unit 11, the determination unit 12, and the display unit 13.

[0062] An example of the operation of the voice analysis device 1 configured as above will be described with reference to the flowchart of FIG.

[0063] The feature extraction unit 11 receives the voice data Vd and the gender information Gi (step S10) and converts the voice data Vd into amplitude data by Fourier transform (step S20). Next, the feature extraction unit 11 calculates the average value of the amplitude data included in each channel range in each scale range of each octave range as the channel amplitude (step S30). Then, using the maximum channel amplitude in each octave range, the feature extraction unit 11 calculates the channel amplitude ratio from the channel amplitude of that channel range in each channel range in each scale range of each octave range (step S40). At this time, high-pass filtering may be performed.

[0064] The feature extraction unit 11 integrates the channel amplitude ratios included in each scale range of each octave range to obtain a scale integral value (step S50). Then, the feature extraction unit 11 determines the gender of the user from the gender information Gi, and inputs the conscious octave information Coi and adjustment value information Nvi corresponding to the determined gender from the consciousness correspondence information table stored in the storage unit 14. The feature extraction unit 11 confirms the octave ranges corresponding to the conscious consciousness octave range, preconscious octave range, and latent octave range from the conscious octave information Coi (step S60), and performs normalization processing on the scale integral value in each conscious octave range using the adjustment value included in the adjustment value information Nvi to obtain a scale feature amount (step S70). The feature extraction unit 11 outputs the scale feature amount in the conscious octave range as a thinking octave feature Tof, the scale feature amount in the preconscious octave range as a habit octave feature Hof, and the scale feature amount in the subconscious octave range as a quality octave feature Nof to the judgment unit 12.

[0065] In the judgment unit 12, the characteristic analysis unit 121 inputs the thinking octave feature Tof, the habit octave feature Hof, and the quality octave feature Nof, reads out the scale information Si1 from the scale information table stored in the storage unit 14, and performs characteristic analysis processing by associating them to obtain the individuality information Pi (step S80). The individuality information Pi is output to the display unit 13.

[0066] The type determination unit 122 performs type determination processing using the quality octave feature Nof (step S90), and reads out the features of the determined type as type information Ti from a type information table stored in the storage unit 14 (step S100). The type information Ti is output to the display unit 13.

[0067] The display unit 13 displays the scale features in a format as shown in Fig. 6 and / or Fig. 7 based on the personality information Pi (step S110). Furthermore, the display unit 13 displays the personality type of the user and its features in a format as shown in Fig. 8 based on the type information Ti (step S120). The order of data input and calculations in FIG. 9 can be changed as appropriate.

[0068] The following modifications are possible to this embodiment.

[0069] Although the voice analysis device 1 uses a voice in which the user's name is spoken as voice data, voice data in which something other than the user's name is spoken may also be used. For example, voice data in which the name of an object other than the user's own, such as another person's name or a company name, is spoken may also be used. In this case, because the spoken voice expresses emotions and feelings toward the object, differences may occur in the characteristics of the characteristic elements and types compared to when the user's name is spoken.

[0070] In the voice analysis device 1, the judgment unit 12 judges the user's personality using the scale features in the conscious octave range, the preconscious octave range, and the latent octave range, and does not use the scale features in other octave ranges. However, the scale features in other octave ranges may be used to judge the user's personality. Conversely, the feature extraction unit 11 may perform the process of determining the scale features only for the conscious octave range. Furthermore, although the octave ranges associated with each consciousness are changed by gender, they may also be changed by factors other than gender, such as age or race.

[0071] In the voice analysis device 1, the display unit 13 may perform the characteristic analysis process of the characteristic analysis unit 121 in the judgment unit 12. That is, the display unit 13 inputs the thinking octave feature Tof, the habit octave feature Hof, and the quality octave feature Nof, reads out the scale information Si1 from the scale information table stored in the storage unit 14, and displays them. In this case, the judgment unit 12 may include only the type judgment unit 122, and may input only the quality octave feature Nof. Alternatively, the type determination unit 122 in the determination unit 12 may output only the determined type to the display unit 13, and the display unit 13 may read out the characteristics of the type corresponding to the input type from a type information table stored in the memory unit 14 and display it.

[0072] The speech analysis device 1 may store the thinking octave features Tof, habit octave features Hof, and quality octave features Nof output by the feature extraction unit 11, for example, in the storage unit 14. By storing these data obtained in the past, it is possible to check changes in the user's personality over time. To check changes over time, the speech analysis device 1 may store the user's voice data Vd and / or type information Ti. When the speech analysis device 1 stores data for multiple users, for example, an identification number or the like that identifies each user may be stored together with the data so that the user can be identified.

[0073] Another embodiment of the present invention will now be described.

[0074] First, a second embodiment of the present invention will be described. In the voice analysis device 1 of the first embodiment, the judgment unit 12 judges the user's personality and type based on characteristic elements, but in addition to these, it is also possible to obtain the ability development level, which indicates how well the user is demonstrating their abilities.

[0075] An example of the configuration of a voice analysis device 1 (second embodiment) that also calculates the level of ability performance is shown in Fig. 10. Compared to the voice analysis device 1 in the first embodiment shown in Fig. 1, in the voice analysis device 2 in the second embodiment, the determination unit 12 and the display unit 13 are changed to a determination unit 22 and a display unit 23, respectively.

[0076] The judgment unit 22 performs characteristic analysis processing and type judgment processing, and also determines an ability development level, similar to the judgment unit 12 in the voice analysis device 1. An example of the configuration of the judgment unit 22 is shown in Fig. 11. Compared to the judgment unit 12 shown in Fig. 5, the judgment unit 22 additionally includes a development level determination unit 123. The thinking octave feature Tof and the talent octave feature Nof are also input to the development level determination unit 123.

[0077] The performance level determination unit 123 determines the ability performance level of the user. In the voice analysis device 2, it is considered that an ability is being demonstrated when what exists in the subconscious, which is the essential quality domain, is expressed in the conscious, which is the thinking domain, and the degree of ability demonstration can be measured by the gap between the conscious and subconscious. The voice analysis device 2 defines the degree of ability demonstration based on information regarding the difference between the scale feature in the octave range of the conscious mind and the scale feature in the octave range of the subconscious mind. For example, for each scale, the degree of ability demonstration is defined based on the number and magnitude of differences that result in negative values, using the result (difference) obtained by subtracting the scale feature in the octave range of the subconscious mind from the scale feature in the octave range of the conscious mind. For example, the degree of ability demonstration is defined as 100 when there are no differences that result in negative values, and as 10 when there are five differences that result in values ​​of -5 or less.

[0078] The performance level determination unit 123 receives the thinking octave features Tof and the quality octave features Nof output from the feature extraction unit 11. Then, the performance level determination unit 123 calculates, for each scale, the difference between the scale feature amount included in the thinking octave features Tof and the scale feature amount included in the quality octave features Nof, and determines the ability performance level using the calculated difference. The determined ability performance level is output to the display unit 23 as an ability performance level Ad.

[0079] The display unit 23 receives the personality information Pi and the type information Ti, and displays the scale feature values ​​for each of the three consciousnesses and the personality type of the user, similar to the display unit 13 in the voice analysis device 1. Furthermore, the display unit 23 receives the ability exertion level Ad, and displays the ability exertion level of the user.

[0080] The operation example of the voice analysis device 2 is the same as the operation example of the voice analysis device 1, except that the above operation by the performance level determination unit 123 and the operation of displaying the ability performance level on the display unit 23 are added.

[0081] A third embodiment of the present invention will now be described. Although the voice analysis device 1 in the first embodiment judges a personality from a user's voice data by itself, it is also possible to judge a user's personality using a voice analysis system that includes, for example, a terminal device that transmits the user's voice data and receives the personality judgment results, and a voice analysis device that receives the user's voice data and transmits the personality judgment results. By connecting the terminal device and the voice analysis device via a network, a user can judge their personality from any location.

[0082] FIG. 12 shows an example of the configuration of the above-described voice analysis system (third embodiment). The voice analysis system 3 includes a terminal device 4 used by a user and a voice analysis device 5 that determines the user's personality, which are connected via a network 6 formed by the Internet or the like. The terminal device 4 may be a general-purpose information device with communication capabilities, such as a smartphone, tablet, or personal computer, and the transmission of voice data and the reception of the determination results are performed by a program (application program). The voice analysis device 5 may be a general-purpose server, for example, and the user's personality is determined by a program running on the server. Note that the terminal device 4 and / or the voice analysis device 5 may be dedicated devices rather than general-purpose devices, and some or all of the processing may be performed by hardware.

[0083] The terminal device 4 inputs the user's voice using a voice input means such as a microphone, and transmits the voice data Vdc to the voice analysis device 5 via communication means connected to the network 6 wirelessly or via a wired connection. Furthermore, information indicating the user's gender is input using a data input means such as a touch panel, and similarly transmitted to the voice analysis device 5 as gender information Gic via communication means. If the voice analysis device 5 can connect to multiple terminal devices 4, the terminal device 4 transmits, for example, an identification number or the like (hereinafter referred to as the "terminal number") that identifies the terminal device 4 to the voice analysis device 5 so that the terminal device 4 that transmitted the voice data Vdc or gender information Gic can be identified. Alternatively, the user's voice may be recorded using a device other than the terminal device 4, and the terminal device 4 may transmit the recorded voice data Vdc to the voice analysis device 5. Alternatively, the user's gender or the like may be registered in advance in the voice analysis device 5, and an identification number or the like (hereinafter referred to as the "user number") that identifies the user may be transmitted from the terminal device 4 to the voice analysis device 5, and the voice analysis device 5 may determine the user's gender based on the user number. In this case, if the terminal device 4 can be identified by the user number, it is not necessary to transmit the terminal number from the terminal device 4 to the voice analysis device 5. The terminal device 4 may transmit these pieces of data together as one piece of data to the voice analysis device 5.

[0084] An example of the configuration of the voice analysis device 5 is shown in Fig. 13. Compared to the voice analysis device 1 in the first embodiment shown in Fig. 1, the voice analysis device 5 in the third embodiment does not include a display unit 13 but newly includes a communication unit 15. The feature extraction unit 11 inputs voice data Vd and gender information Gi from the communication unit 15, and the judgment unit 12 outputs personality information Pi and type information Ti to the communication unit 15.

[0085] The communication unit 15 receives the voice data Vdc and the gender information Gic transmitted from the terminal device 4 via the communication means provided in the voice analysis device 5, and outputs them as the voice data Vd and the gender information Gi, respectively, to the feature extraction unit 11. At this time, if information necessary for transmission has been added to the voice data Vdc and the gender information Gic, they may be output without this information as the voice data Vd and the gender information Gi.

[0086] Furthermore, the communication unit 15 receives the individuality information Pi and type information Ti output by the determination unit 12 and transmits them as individuality information Pic and type information Tic via the communication means, respectively. At this time, if it is necessary to add information required for transmission (such as a terminal number), the information may be added to the individuality information Pi and type information Ti and transmitted as the individuality information Pic and type information Tic. Note that the communication unit 15 may compile the individuality information Pic and type information Tic into one piece of data and transmit it to the terminal device 4.

[0087] The terminal device 4 receives, via a communication means, the individuality information Pic and type information Tic transmitted from the voice analysis device 5. Then, similar to the display unit 13 of the voice analysis device 1, the result of the voice analysis device 5's judgment of the user's individuality based on the individuality information Pic and type information Tic is displayed on a display means such as a liquid crystal panel. At this time, the voice analysis device 5 may create images such as a graph as shown in Fig. 6 or a radar chart as shown in Fig. 7, and transmit information about these images to the terminal device 4 as the individuality information Pic and type information Tic, and the terminal device 4 may only display the transmitted images.

[0088] The operation example of the voice analysis device 5 is the same as the operation example of the voice analysis device 1 except that the above-mentioned operation by the communication unit 15 is added and there is no operation by the display unit 13.

[0089] The speech analysis device 5 may also be provided with a display unit 13, similar to the speech analysis device 1, so that the speech analysis device 5 can also display the judgment results of the user's personality. In this case, the terminal device 4 may only transmit the user's voice data, etc. Furthermore, the terminal device 4 executes an application program to transmit and receive data to and from the speech analysis device 5. Alternatively, a website for judging the user's personality may be constructed on the speech analysis device 5, a general-purpose web browser may be installed on the terminal device 4, and information such as HTML (Hypertext Markup Language) pages may be transmitted and received between the website on the speech analysis device 5 and the web browser on the terminal device 4 via the network 6 using HTTP (Hypertext Transfer Protocol) as the network protocol, thereby transmitting and receiving data to and from the speech analysis device 5.

[0090] The above-described embodiment can be realized by a computer and memory configuration, with the storage unit 14 as a memory and the processing of the other components implemented as a program as described above. Each component can also be implemented as hardware, such as a dedicated integrated circuit (IC) or field programmable gate array (FPGA). Furthermore, while the above-described embodiment has been described in the form of an apparatus or system, the present invention can also be implemented in the form of a method or program.

[0091] The present invention is not limited to the above-described embodiments, and various modifications are possible within the scope of the present invention. Furthermore, matters not explicitly disclosed in the above-described embodiments do not deviate from the scope of ordinary practice by a person skilled in the art, and values ​​that can be easily assumed by a person skilled in the art can be adopted. [Explanation of symbols]

[0092] 1, 2, 5 Voice analyzer 3. Voice analysis system 4 Terminal Devices 6 Network 11 Feature Extraction Unit 12, 22 Judgment section 13, 23 Display section 14 Storage section 15 Communications Department 121 Characteristic Analysis Department 122 Type Judgment Department 123 Performance Determination Unit

Claims

1. A voice analysis device that performs analysis based on user voice data, a feature extraction unit that extracts a scale feature, which is a feature for each musical scale in an octave range, which is a range in units of an octave, by frequency analysis of the audio data; a judgment unit that associates characteristic elements defined based on judgment criteria, which are criteria based on sensations, and behavioral criteria, which are criteria based on viewpoints, with musical scales, and judges the personality of the user based on musical scale features in the conscious octave range, which is an octave range related to the conscious mind, the preconscious octave range, which is an octave range related to the preconscious mind, and the subconscious octave range, which is an octave range related to the subconscious mind.

2. The voice analysis device according to claim 1 , wherein the determining unit determines the individuality of the user by determining the scale feature extracted by the feature extracting unit using a threshold value for the scale feature.

3. 3. The voice analysis device according to claim 2, wherein the judgment unit defines a plurality of types based on the judgment criteria and the behavioral criteria, and determines the type of the user by using the threshold value to judge the scale features of the subconscious octave range extracted by the feature extraction unit.

4. 2. The speech analysis device according to claim 1, wherein the judgment unit defines an ability demonstration level based on information relating to a difference between the scale feature values ​​of the conscious octave range and the scale feature values ​​of the subconscious octave range, and determines the ability demonstration level of the user based on the scale feature values ​​of the conscious octave range and the scale feature values ​​of the subconscious octave range extracted by the feature extraction unit.

5. The voice analysis device according to claim 1 , further comprising a display unit that displays a result of the determination by the determination unit.

6. The voice analysis device according to claim 5 , wherein the display unit displays the scale feature extracted by the feature extracting unit for each characteristic element.

7. The speech analysis device according to claim 6 , wherein the display unit displays the scale features using colors defined for each scale.

8. 8. The voice analysis device according to claim 1, wherein the voice data is data in which the user speaks his or her own name.

9. A speech analysis system comprising the speech analysis device according to any one of claims 1 to 4 and a terminal device used by the user, the terminal device transmits the user's voice data to the voice analysis device; A voice analysis system characterized in that the voice analysis device transmits the result of the judgment by the judgment unit to the terminal device.

10. A voice analysis method for performing analysis based on user voice data, comprising: a feature extraction step of extracting a scale feature, which is a feature for each musical scale in an octave range, which is a range in units of an octave, by frequency analysis of the audio data; and a judgment step of associating characteristic elements defined based on judgment criteria, which are criteria based on sensations, and behavioral criteria, which are criteria based on viewpoints, with musical scales, and judging the personality of the user based on musical scale features in the conscious octave range, which is an octave range related to the conscious mind, the preconscious octave range, which is an octave range related to the preconscious mind, and the subconscious octave range, which is an octave range related to the subconscious mind.

11. The voice analysis method according to claim 10, wherein the voice data is data in which the user speaks his or her own name.

12. 11. The voice analysis method according to claim 10, wherein the determining step determines the personality of the user for the purpose of providing counseling to the user.

13. The voice analysis method according to claim 12 , wherein the determining step determines the personality of the user for the purpose of counseling including advice on health and beauty for the user.

14. The voice analysis method according to claim 10, wherein the ability exertion level of the user is calculated based on information about a difference between the scale feature amount of the conscious octave range and the scale feature amount of the subconscious octave range.

15. A speech analysis program for causing a computer to execute the speech analysis method according to any one of claims 10 to 14.

Images