JP2006346471A - Sound generation method, computer readable storage medium, stand-alone sound generation and playback device, and network communicative sound generation and playback system - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a sound generation method to basic information having chaos and fractal, by calculating characteristics of chaos and fractal from the basic information having chaos and fractal, and by applying a generation rule to the obtained data of chaos and fractal. <P>SOLUTION: This sound generation method includes: a basic information converting process, which converts basic information having chaos and fractal into data that are numerically operable; a chaotic space generating process, which calculates a chaos attractor and a fractal feature on the basis of the data converted by the basic information converting process; a fractal space generating process, which generates a fractal space, and a sound generating process, which generates a sound file in compliance with a predetermined sound generation rule, from the data in the chaotic space and the fractal space generated by the chaotic space generating process and the fractal space generating process. <P>COPYRIGHT: (C)2007,JPO&INPIT

Description

  The technical field of the present invention generates chaotic and / or fractal sound that is useful for the human body by utilizing the basic information having chaotic and / or fractal characteristics in the sound generation method. It relates to a sound generation method. The present invention also relates to a computer-readable storage medium that records a sound generation program based on the sound generation method, a stand-alone type sound generation / reproduction device and a network distribution type sound generation / reproduction system that generate a sound that matches the purpose. .

  In the conventional sound generation method, there are many compact discs and the like that record music on the theme of “healing” such as so-called healing music. As examples, there are known α-wave music in which α-waves of brain waves are likely to appear due to the intentions of composers and arrangers, healing sounds that record natural sounds as they are, and the like.

  On the other hand, in recent years, a nonlinear processing method for performing nonlinear processing on a biological signal of a human body by chaos theory has been attracting attention. A technique for diagnosing a health condition based on whether or not a pulse wave and a heartbeat signal satisfy a condition that they are chaos has been disclosed (for example, see Patent Document 1 below). This is based on the knowledge that pulse waves and heartbeat signals obtained from a healthy living body are chaotic. In order for some data to be chaotic, conditions such as having a non-integer fractal dimension and a positive maximum Lyapunov number are generally used. Similarly, as a technique for objectively diagnosing a person's mind and body state, a technique using a chaos attractor or a Lyapunov exponent is disclosed (for example, see Patent Document 2 below).

  One of the methods for quantitatively evaluating the human mind and body condition is electroencephalogram analysis. In general, it is said that α waves are emitted when relaxing, and β waves are emitted when excited. As a technique for evaluating a human mental state, a technique for evaluating a human mental state by obtaining an average power value in two types of electroencephalogram bands and calculating a ratio is disclosed (for example, Patent Document 3 below). reference.).

JP-A-4-208136 (page 2-8) JP-A-6-54813 (page 2-6) JP 2002-577 A (page 3-5, FIG. 5)

  However, in the conventional sound generation method, since the sound generation intended by the composer and the arranger is generated, it is not possible to generate a sound based on chaotic basic information such as personal biometric information. It was.

  An object of the present invention is to calculate chaos and / or fractal characteristics of basic information having chaotic properties and / or fractal properties by converting basic information having chaotic properties and / or fractal properties into numerically operable data. The object of the present invention is to provide a sound generation method for basic information having chaos and / or fractal by applying a generation rule to the obtained chaos and / or fractal data.

  It is another object of the present invention to provide a sound generation method capable of making a more comfortable sound while feeding back an evaluation result for the obtained sound.

  In addition, a computer-readable storage medium in which a sound generation program according to the above method is recorded, a measurement of a biological signal from an individual information provider, and a stand-alone sound generation / reproduction device that reproduces the generated sound, and a network distribution type sound generation / reproduction System.

The present invention is as follows.
1. A basic information conversion procedure for converting basic information having chaos into data that can be numerically calculated, and a chaos characteristic space generation procedure for generating a chaos characteristic space by calculating a chaos attractor based on the data converted by the basic information conversion procedure And a sound generation procedure for generating a sound file from data in the chaos characteristic space generated by the chaos characteristic space generation procedure according to a predetermined sound generation rule.
2. Basic information conversion procedure for converting basic information having fractal properties into data that can be numerically calculated, and fractal characteristic space generation for generating fractal characteristic space by extracting fractal features based on the data converted by the basic information conversion procedure A sound generation method comprising: a procedure; and a sound generation procedure for generating a sound file from a fractal characteristic space generated in the fractal characteristic space generation procedure according to a predetermined sound generation rule.
3. A biological signal conversion procedure that converts a time-series generated signal of an information provider individual into data that can be numerically calculated, and a chaos attractor is calculated based on the data converted by the biological signal conversion procedure to generate a chaos characteristic space A chaos characteristic space generation procedure, and a sound generation procedure for generating a sound file adapted to the information provider from data in the chaos characteristic space generated in the chaos characteristic space generation procedure according to a predetermined sound generation rule. Sound generation method characterized by.
4). A biosignal conversion procedure for converting a time-series generated signal of an information provider individual into data that can be numerically calculated, and a self-similar feature is extracted based on the data converted by the biosignal conversion procedure, thereby creating a fractal characteristic space. A fractal characteristic space generation procedure to be generated, and a sound generation procedure to generate a sound file adapted to the information provider according to a predetermined sound generation rule from the fractal characteristic space generated by the fractal characteristic space generation procedure. A characteristic sound generation method.
5. The biological signal conversion procedure includes a biological signal measurement procedure for measuring a biological signal, a frequency analysis procedure for calculating biological signal data measured in the biological signal measurement procedure as numerical data for each frequency band, and the frequency analysis procedure. And 3. a sound generation procedure corresponding to a nerve type characteristic in the living body of the individual information provider. The described sound generation method.
6). The biological signal conversion procedure includes a biological signal measurement procedure for measuring a biological signal, a frequency analysis procedure for calculating biological signal data measured in the biological signal measurement procedure as numerical data for each frequency band, and the frequency analysis procedure. And 4. a sound generation procedure corresponding to a nerve type characteristic in the living body of the individual information provider. The described sound generation method.
7). The chaos characteristic space generation procedure is a state evaluation procedure that compares the numerical data of the nerve type characteristics in the living body of the information provider individual calculated in the frequency analysis procedure, and evaluates the psychosomatic state of the information provider, and the state 2. A cutting plane changing procedure for changing a plane for cutting the chaotic attractor according to the evaluation in the evaluation procedure. The described sound generation method.
8). The fractal characteristic space generation procedure compares the numerical data of nerve type characteristics in the living body of the information provider individual calculated in the frequency analysis procedure, and evaluates the state and condition of the information provider, and the state 3. A scaling width changing procedure for changing a scaling width for extracting a fractal feature in accordance with the evaluation in the evaluation procedure. The described sound generation method.
9. The sound generation procedure includes an interface capable of interacting with an information provider of the biological signal, a condition input procedure that allows a condition relating to sound generation to be input, and the sound generation rule according to the condition input in the condition input procedure Generating a sound file according to the generation rule set in the generation rule setting procedure. The described sound generation method.
10. The sound generation procedure includes an interface capable of interacting with an information provider of the biological signal, a condition input procedure that allows a condition relating to sound generation to be input, and the sound generation rule according to the condition input in the condition input procedure Generating the sound file according to the generation rule set in the generation rule setting procedure. The described sound generation method.
11. In the state evaluation procedure, the psychosomatic state is evaluated based on the appearance rate of the α wave of the electroencephalogram. The described sound generation method.
12 In the state evaluation procedure, the psychosomatic state is evaluated based on the appearance rate of α waves of the electroencephalogram. The described sound generation method.
13. 2. At least one of pulse wave, electrocardiogram, electroencephalogram, myoelectricity, and respiration is used as the biological signal. To 10. The sound generation method according to any one of the above.
14 Above 1. To 12. A computer-readable storage medium having recorded thereon a program for causing a computer to execute at least one of the sound generation methods according to any one of the above.
15. 13. above. A computer-readable storage medium in which a program for causing a computer to execute the sound generation method described in 1 is recorded.
16. Means for measuring a biological signal; To 12. A computer that executes at least one of the sound generation methods according to any one of the above, a means for reproducing the sound generated by the sound generation method, and information for providing the biological signal and listening to the sound A stand-alone type sound generating and reproducing apparatus comprising: means for measuring a provider's individual state.
17. Means for measuring a biological signal; A computer that executes the sound generation method according to claim 1, means for reproducing the sound generated by the sound generation method, means for measuring the state of an individual information provider who provides the biological signal and listens to the sound, A stand-alone type sound generating and reproducing device comprising:
18. Above 1. To 12. A server computer that executes at least one of the sound generation methods according to any one of the above, a computer network that includes means for measuring biological signals necessary for the sound generation method and means for reproducing sound A network-distributed sound generating / reproducing system, comprising:
19. 13. above. A server computer that executes the sound generation method according to claim 1, a means for measuring a biological signal necessary for the sound generation method, and a means for executing the remote computer connected by a computer network including means for reproducing the sound. A network-distributed sound generation / playback system comprising:

  According to the sound generation method of the present invention, chaos and / or fractal of basic information having chaos and / or fractal is obtained by converting basic information having chaos and / or fractal to data that can be numerically calculated. By calculating the characteristics and applying the generation rules to the obtained chaos and / or fractal data, it is possible to generate sound for the basic information having chaos and fractal characteristics.

  According to the sound generation method of the present invention, by measuring the pulse wave and brain wave of the information provider, the plane for cutting the chaotic attractor and the scaling width for extracting the fractal feature are changed, thereby making it more useful for the information provider. It is possible to generate a sound that makes it possible to obtain a peaceful effect.

In addition, when changing the sound generation rule while interacting with the information provider, it is possible to generate a sound that provides a more relaxing effect for the information provider.
Furthermore, when using biological information as basic information for generating a sound file, it is possible to generate a sound adapted to the individual of the biological information provider. For example, the biological information provider when the mood or physical condition of the biological information provider is good Based on the information, it is possible to generate a sound that exhibits effects such as improvement of physical condition, recovery of mood, etc., when the physical condition is about to be lost or when the person is feeling sick. Similar to being able to calm an infant by letting the infant hear the sound that he had heard in his mother's womb before birth, he can heal or encourage the person by listening to individual sounds. The nature of the sound itself is different from listening to the sound of the natural world that is generally said to heal.

  What is considered to be the best by evaluating the psychosomatic state based on the data for each frequency band of the biological signal and repeating the change and evaluation of the cut plane and scaling width in order to converge the obtained evaluation value to a more favorable state Will be obtained.

  According to the computer readable storage medium storing the program for the sound generation method, the sound generation method can be executed to generate a sound.

  According to the stand-alone type sound generation / playback apparatus, it is possible to generate a sound using the sound generation method by using only the apparatus itself.

  According to this network distribution type sound generation and reproduction system, sound can be generated on the server computer side using the sound generation method based on a biological signal obtained by a remote computer.

Hereinafter, embodiments of the present invention will be described with reference to the drawings.
The “chaotic property” mentioned above refers to a property that does not have periodicity or almost periodicity even in a system that follows a deterministic equation, and exhibits seemingly irregular behavior. Further, the “fractal property” means having self-similarity.

  The “basic information” having chaotic properties and fractal properties is a signal derived from living things and weather. Examples of this include biological signals, as well as information that occurs in nature, such as sound data such as river buzz, measurement data of wind intensity and / or direction, vibration data measured by a seismometer, etc. Can do.

  The “biological signal” may be any signal that can be measured from the body and can be arbitrarily selected. Examples of this include an electroencephalogram, a pulse wave obtained by measuring a pulse, an electrocardiogram that is a temporal change in action potential associated with the heartbeat, a myoelectric that is a temporal change in potential associated with muscle contraction, and the mouth. Exhalation which is a right change such as the pressure or flow rate of the breath exhaled from the air can be mentioned.

  The “sound file” is not particularly limited as long as it can be reproduced as sound. Examples of this include a format in which audio signals are sampled in a format such as PCM (Pulse Code Modulation), and a format in which sound frequencies and lengths are listed in a predetermined format. More specifically, a standard MIDI format file (also referred to as MIDI file or SMF file) or a file sampled in a format such as PCM (WAVE file) can be used.

  As shown in FIG. 1, the generation method of this embodiment measures basic information A having chaotic properties and fractal properties and converts it into data that can be numerically calculated, and calculates a chaotic attractor. The chaos characteristic space generation procedure 2 for generating the chaos characteristic space obtained by the above and the sound generation procedure 3 for generating the sound file C that performs chaotic behavior by applying the sound generation rule B are provided.

  The basic information conversion unit 1 converts the basic information A having chaotic properties and / or fractal properties into data that can be numerically calculated by the information measuring device 101, and the time series chaotic properties and fractal properties are obtained by the measuring computer 102. The basic information conversion procedure for measuring the basic information A is provided.

  In addition, the sound generation method of this embodiment includes a basic information conversion unit 1 that measures basic information A having chaos and fractal properties and converts it into numerically operable data, as shown in FIGS. A sound generation procedure 3 for generating a sound file C having a fractal behavior by applying the fractal characteristic space 6 obtained by extracting the fractal feature and the sound generation rule B is provided.

  A computer-readable storage medium storing a sound generation method program using a biological signal of the present invention is obtained by a program for causing a computer to generate a sound using a biological signal, a biological signal used for sound generation, and the sound generation. And a sound file.

  The sound generation / playback apparatus of the present invention is configured to include measurement of a biological signal and playback of a sound file in an integrated type or a network connection type.

  As shown in FIG. 3, the integrated sound generating / reproducing apparatus includes a listener state measuring unit, a biological signal measuring unit, a sound generating unit, and a sound reproducing unit.

  Further, as shown in FIG. 4, the network type sound generation / playback apparatus includes a server computer including a sound generation unit and a client terminal which is a remote computer connected to the server computer via a network. The client terminal includes a listener state measurement unit, a biological signal measurement unit, and a sound reproduction unit. Further, the network transmits the listener state and the biological signal obtained by the listener state measurement unit and the biological signal measurement unit to the sound generation unit of the server computer, and the sound file obtained from the sound generation unit of the server computer. Used to transmit to the sound playback unit of the client terminal.

  The “listener state measurement unit” measures a listener state used in cutting plane learning. The listener state can be obtained by an arbitrary measurement method, for example, by measuring an electroencephalogram. The listener state measurement unit measures brain waves and obtains the listener state.

  The “biological signal measurement unit” measures a biological signal. Moreover, a pulse wave is used as a biological signal.

  The “sound generation unit” includes a computer capable of executing a sound generation method, and generates sound based on the listener state obtained by the listener state measurement unit and the biological signal obtained by the biological signal measurement unit. To do.

  The “sound reproduction unit” includes a speaker, an amplifier, and the like that reproduce the sound created by the sound generation unit.

The “network” may be of any type and form.
Note that the computer-readable storage medium and the sound generation / playback apparatus in which the sound generation method program is recorded may use a sound generation method using either chaotic property or fractal property, or both sound generation methods. It is also possible to select and use arbitrarily. Furthermore, sound files obtained using a plurality of sound generation methods can be mixed into one.

1. Sound generation from chaotic basic information The chaotic characteristic space generating procedure 2 regards the measured basic information A having time-series chaotic properties as one-dimensional time-series data and defines x (t). Here, t represents the passage of time. FIG. 5 and FIG. 6 show pulse wave graphs as an example of the measured basic information A having time-series chaos.

  The chaos attractor calculation unit 201 calculates a chaos attractor by using two arbitrary parameters, a time delay τ (second) and an embedding dimension m (dimension). Chaotic attractor is m-dimensional multivariate data. The formula for defining the chaos attractor is shown in Equation 1. In addition, two examples of the three-dimensional graph of the chaotic attractor calculated using Equation 1 with the time delay set to 0.1 second and the embedding dimension set to three dimensions for the measured pulse wave are shown in FIGS. It is shown in FIG.

  The cut surface constituting unit 202 calculates a small shift of the chaotic period using the Poincare map for the calculated chaotic attractor. In this calculation, the solution orbit of the chaotic attractor is cut along a predetermined plane, and an (m−1) -dimensional point sequence p (i) plotted on the cut plane (Poincare cut plane) is obtained. Here, i is the time transition order in the set of intersections.

Reference point setting unit 203, a reference point P _b to be used to determine the vector with the intersection of the chaotic attractor and the cutting plane is set at a predetermined position on the cut surface. This reference point P _b may be arbitrarily selected position.

Rule application unit 301 of the sound generation step 3, the difference between the position vector p _b of the position vector p (i) and the reference point p _b of the chaotic sequence of points is the intersection of the cutting plane of the tractor p (i) P (i) Ask for. For example, it can be calculated as a vector P (i) between the reference point as shown in FIG. 9 p _b the intersection of the cutting plane p (1) ~p (5) .

  Sound generation is performed by combining the vector P (i) and the sound generation rule B. In sound generation, the minimum unit (phoneme) of sound composed of height n, length l, and strength v is determined. A sound generated from an m-dimensional chaotic attractor can be composed of a sequence of at least m chords. The sound generation rule B is given by, for example, a conversion formula expressed by Equations 2 to 5 from the vector P (i) of the point sequence p (i) on the Poincare section.

Here, S _k (i) is the i-th phoneme in channel k. M (a, b) is a scale element having an octave height a and a scale number b, and r is the number of sounds in the selected scale. Furthermore, P _k (i) represents the k-th element value in the vector P (i), and mod represents the remainder operation.

  Note that rules other than those described above can be applied to the sound generation rule. For example, the deflection angle of the two vectors P (i−1) and P (i) obtained from the previous and current intersection points on the Poincare cut surface. Any element such as a length and an angle that can be obtained from the intersection and the reference point can be used. Furthermore, using these elements, it is possible to determine performance effects such as the type of musical instrument (tone color), chord progression, vibrato, portamento, and echo, and generate a sound with a higher psychological effect.

2. Sound generation from fractal basic information As a method for obtaining the fractal characteristic space 6, a method for calculating an attractor from each basic information obtained by repeatedly performing scaling conversion on the measured basic information having fractal characteristics in time series, and And a method of directly obtaining from the measured basic information.

  The fractal characteristic space 6 regards the measured basic information A having time-series fractal properties as one-dimensional time-series data and defines x (t). Here, t represents the passage of time. Two examples of such x (t) are shown in FIGS. 10 to 13 and FIGS. 14 (a), 14 (b), 15 (a), and 15 (b). . FIGS. 11 to 13 are enlarged views of a part of FIG. 10. FIGS. 14 (b), 15 (a), and 15 (b) are those of FIG. 14 (a). A part of it is enlarged. Such x (t) is obtained by performing scaling conversion on a pulse wave graph, which is an example of the basic information A having time series fractal characteristics, and extracting self-similar fractal features. .

  For example, FIG. 13 shows twice the operation of extending the time series direction by a factor of 1/5 (0.8 seconds) of the time series data of the fingertip volume pulse wave measured for 4 seconds by the pulse wave sensor. FIG. 16 to FIG. 19 show the three-dimensional graph of the attractor (b) calculated using Equation 1 from each time-series pulse wave (a).

  By mixing the sound sequence to which the sound conversion rule is applied to the three attractor spaces obtained in this way, a sound for basic information having fractal characteristics is generated.

  It is also possible to extract the fractal features directly from the measured basic information by observing the complex shape caused by the change of time series data as the variation of declination and observing the variation of declination at multiple appropriate scales. Is possible.

For example, with respect to the measured fingertip plethysmogram shown in FIG. 20 (a), the deviation θ (t) in the range | t _i − Measure while changing t _{i ± k} |. In this method, a point P _i (t _i , y (t _i )) of basic information is determined and used as a reference in a predetermined cycle, and the previous period P _i−k (t _i−k , y (t _i−k )) is used. ) And the next period P _{i + k} (t _{i + k} , y (t _{i + k} )) is _defined as θ (t _i ).

By changing the range | t _i −t _{i ± k} | from 0.5 to 0.005 (seconds), as shown in FIGS. 20 (a), 21 (a) and 21 (b). Three variations of declination θ (t) are obtained. This series θ (t) is replaced with p (i) in the sound conversion rule, and a sound series is generated at a timing that matches the rhythm of the pulse. By mixing the obtained three sounds, a sound for basic information having fractal characteristics is generated.

3. Cut surface learning As shown in FIG. 23, the cut surface learning 4 includes an electroencephalogram measurement unit 401 that measures the brain wave of the information provider, a frequency analysis unit 402 that calculates the electroencephalogram data as numerical data for each frequency band, It is configured to include an electroencephalogram evaluation unit 403 that compares numerical data for each band and evaluates a psychosomatic state, and a cutting plane changing unit 404 that changes a plane that cuts the chaotic attractor according to the evaluation.

  The cut surface learning 4 reproduces a sound generated based on basic information such as an electroencephalogram, as in the flowchart explaining the cut surface learning process shown in FIG. Next, a brain wave of a person who listens to the reproduced sound is detected and evaluated, and feedback for changing the cut surface according to the evaluation is performed. As a result, a more suitable sound can be generated.

  In addition, as shown in FIG. 25, the scaling width learning 7 in the present embodiment includes an electroencephalogram measurement unit 701 that measures an information provider's electroencephalogram, and a frequency analysis unit 702 that calculates electroencephalogram data as numerical data for each frequency band. And an electroencephalogram evaluation unit 703 that compares numerical data for each frequency band and evaluates a psychosomatic state, and a scaling width change unit 704 that changes a scaling width for extracting a fractal feature according to the evaluation. The

4). Scaling Width Learning The scaling width learning 7 reproduces a sound generated based on basic information such as an electroencephalogram as in the flowchart for explaining the scaling width learning process shown in FIG. Next, the brain wave of the person who listened to the reproduced sound is detected and evaluated, and feedback is performed to change the scaling width according to the evaluation. As a result, a more suitable sound can be generated.

The electroencephalogram measurement units 401 and 701 measure the electroencephalogram in each part by attaching electrodes to a plurality of information provider's heads, and convert them into data that can be numerically calculated.
In addition, the electroencephalogram measurement method conforms to the International Electroencephalographic Society Standard Method (International 10/20 Method). As electrode positions used for the electroencephalogram measurement, 12 positions (Fp1, Fp2, F7, F8, Fz, C3, C4, Pz, T5, T6, O1, O2) are employed. In addition, finger plethysmogram, eye movement, and electrocardiogram are also measured.

  When measuring an electroencephalogram, for example, as shown in a simplified diagram of an electroencephalogram measurement environment in FIG. 27, a speaker for sound reproduction is installed at a predetermined interval in front of a subject wearing an electroencephalogram measurement electrode. Further, the computer that operates the sound generation method is operated, and the obtained sound file is reproduced by the speaker. Furthermore, as shown in FIG. 28 and FIG. 29, the position where the electrode is mounted is determined according to the electrode position layout diagram according to the International Electroencephalographic Society Standard Method (International 10/20 Method).

  The frequency analysis units 402 and 702 convert the electroencephalogram data obtained by measurement into δ wave (0.5 to 3 Hz), θ wave (3 to 7 Hz), α wave (7 to 13 Hz), β wave (13 to 30 Hz) and γ waves (30 to 40 Hz) are decomposed into frequency bands, and the power spectrum is calculated to obtain numerical data that can grasp the state of the brain, and the brain waves are analyzed.

  The electroencephalogram evaluation unit 403 checks the correlation of numerical data for each frequency band obtained by analysis, evaluates whether the information provider is at rest, and accumulates the result as a reward value. As a measure of peacefulness, the overall content of each frequency band, which is the ratio of waves in each frequency band to the total brain wave, and the spectrum value of each frequency band that represents the strength of the wave in each frequency band, is used as an overall measure. The state of the brain is evaluated by grasping a tendency, and a positive reward value is given to a cut surface with a high evaluation value, and a negative reward value is given to a cut surface with a low evaluation value.

  The electroencephalogram evaluation unit 703 examines the correlation of numerical data for each frequency band obtained by analysis, and evaluates whether the information provider is at ease. As a measure of peacefulness, the content of each frequency band, which is the ratio of waves in each frequency band to the entire brain wave, and the spectrum value of each frequency band that represents the strength of the wave in each frequency band as a whole The brain state is evaluated by grasping a general tendency, and a positive reward value is given to a scaling range where the evaluation value becomes high, and a negative reward value is given to a scaling range where the evaluation value becomes low.

  When the evaluation value becomes low, the cut surface changing unit 404 changes the cut surface in consideration of the reward value accumulated so far and continues the electroencephalogram measurement. When the evaluation value has converged, it is considered that learning has been completed, and the electroencephalogram measurement is terminated with the current cut surface as the optimal cut surface.

  When the evaluation value becomes low, the scaling width changing unit 704 changes the scaling width in consideration of the rewards accumulated so far and continues the electroencephalogram measurement. When the evaluation value has converged, it is assumed that the learning has been completed, and the electroencephalogram measurement is terminated using the current scaling width as the optimal scaling width.

  In the generation rule learning 5 in the present embodiment, as shown in FIG. 30, a user input unit 501 having an interface capable of interacting with an information provider, a generation rule changing unit 502 that changes the sound generation rule B according to the input contents, A sound change unit 503 that generates a sound corresponding to the change is provided. 31 and 32 are flowcharts for explaining the generation rule learning process. As shown in FIGS. 31 and 32, the generation rule learning 5 can change the timbre, volume, etc. of the sound to be generated.

  As a means for evaluating the psychosomatic state of a listener who listens to a sound, a method of investigating the activity of the autonomic nervous system by measuring a biological signal of a pulse or a heartbeat and observing the frequency spectrum of the fluctuation in addition to an electroencephalogram. FIG. 33 (a) is a graph in which the fluctuation of the heart rate measured from the living body at the time of listening to the sound is calculated. The frequency component of this fluctuation is represented by a low frequency component (0.024 to 0.15 (Hz)) and a high frequency. It is divided into two components (0.15 to 0.6 (Hz)), and by observing the high frequency / low frequency which is the ratio of the content rate, the activity of parasympathetic nerve and sympathetic nerve in the living body at the time of sound listening The ratio (referred to as the nerve type characteristic) can be measured, and Fig. 33 (b) is a graph of the analysis result of the nerve type characteristic from the biological signal taking the heart rate variability of Fig. 33 (a) as an example. When it is difficult to measure an electroencephalogram from a sound listener, the value of the high / low frequency value is replaced with an evaluation value of the comfort of the sound listener.

  The user input unit 501 has an interface that allows the information provider to change the musical instrument, the pitch range, the sound intensity range, and the sound length range while listening to the sound. Be able to reflect your preferences.

  The generation rule changing unit 502 applies the item content changed by the information provider using the interface to the generation rule.

  The sound changing unit 503 generates a sound using the changed generation rule.

  In the sound generation method using such a chaotic attractor, as shown in FIG. 1, the basic information conversion unit 1 measures the basic information A of the information provider and converts it into data that can be numerically calculated. Next, a chaos attractor is calculated from this data by the chaos characteristic space generation procedure 2. Thereafter, the obtained chaotic attractor is passed to the sound generation procedure 3 and the sound generation rule B is applied to generate a sound file C that behaves like a chaos.

  In the sound generation method using the fractal feature, as shown in FIG. 2, the basic information conversion unit 1 measures the basic information A of the information provider and converts it into data that can be numerically calculated. Next, a fractal feature is calculated from this data by the fractal characteristic space 6. Thereafter, the obtained fractal feature is passed to the sound generation procedure 3 and a sound file C is generated by applying the sound generation rule B.

It is explanatory drawing for demonstrating the whole structure of the sound production | generation method from the basic information which has the chaotic characteristic of this invention. It is explanatory drawing for demonstrating the whole structure of the sound production | generation method from the basic information which has the fractal characteristic of this invention. It is a block diagram of a stand-alone type sound generating and reproducing apparatus. 1 is a configuration diagram of a network distribution type sound generation / playback apparatus. It is a graph of the measured pulse wave. It is a graph of the measured pulse wave. It is a three-dimensional graph of the calculated chaotic attractor. It is a three-dimensional graph of the calculated chaotic attractor. It is a schematic diagram showing a mode that it is plotted on a cut surface. It is a figure of scaling conversion in the process of fractal feature extraction taking pulse wave data as an example. It is a figure of scaling conversion in the process of fractal feature extraction taking pulse wave data as an example. It is a figure of scaling conversion in the process of fractal feature extraction taking pulse wave data as an example. It is a figure of scaling conversion in the process of fractal feature extraction taking pulse wave data as an example. FIG. 4A is a scaling conversion diagram in the process of fractal feature extraction taking pulse wave data as an example, and FIG. FIG. 14 is an enlarged view of a part of the scaling conversion diagram in the process of fractal feature extraction taking the pulse wave data of FIG. It is a three-dimensional graph (b) of a pulse wave graph (a) and a chaotic attractor calculated by scaling conversion of the pulse wave graph (a). It is a three-dimensional graph (b) of a pulse wave graph (a) and a chaotic attractor calculated by scaling conversion of the pulse wave graph (a). A pulse wave graph (a) and a three-dimensional graph (b) of a chaotic attractor calculated by scaling conversion of the pulse wave graph. FIG. 16B, FIG. 17B, and FIG. 18B, which are three-dimensional graphs of chaotic attractors calculated from each scale-converted pulse wave data, are listed. It is a graph of a fractal feature directly extracted from basic information using pulse wave data as an example, (a) is a time-series data graph of pulse waves, and (b) is | t _i −t _{i ± k} | = 0.5. It is a data graph at the time of second. It is a graph of a fractal feature directly extracted from basic information using pulse wave data as an example, (b) is a data graph when | t _i −t _{i ± k} | = 0.05 seconds, and (b) is | It is a data graph when t _i −t _{i ± k} | = 0.005 seconds. It is a schematic diagram showing the measurement method of the feature extraction in the process of extracting a fractal feature directly from the basic information which made pulse wave data an example. It is description for demonstrating the whole structure of the cut surface learning process of this invention. It is a flowchart explaining the cut surface learning process of this invention. It is explanatory drawing for demonstrating the whole structure of the scaling width learning process of this invention. It is a flowchart explaining the scaling width learning process of this invention. It is a schematic diagram of an electroencephalogram measurement environment. It is the schematic diagram which looked at the electrode position by the International Electroencephalogram Society standard method (international 10/20 method) from the top. It is the schematic diagram which looked at the electrode position by the International Electroencephalographic Society standard method (international 10/20 method) from the front. It is explanatory drawing for demonstrating the whole structure of the production | generation rule learning process of this invention. It is a flowchart explaining the sound production | generation rule learning process from the basic information which has the chaos characteristic of this invention. It is a flowchart explaining the sound production | generation rule learning process from the basic information which has the fractal characteristic of this invention. It is the graph (a) which calculated the fluctuation | variation of the heartbeat measured from the biological body at the time of sound listening, and the graph (b) of the analysis result of the nerve classification characteristic from the biological signal which made the heartbeat fluctuation an example.

Explanation of symbols

  DESCRIPTION OF SYMBOLS 1; Basic information conversion part, 101; Information measuring apparatus, 102; Measurement computer, 2; Chaos characteristic space generation procedure, 201; Chaotic attractor calculation part, 202; Cut surface structure part, 203; Reference point setting part, 3; Sound generation procedure, 4; cutting plane learning, 401; electroencephalogram measurement section, 402; frequency analysis section, 403; electroencephalogram evaluation section, 404; cutting plane changing section, 5; generation rule learning, 501; user input section, 502; Part 503; sound change part 6; fractal characteristic space 7; scaling width learning 701; electroencephalogram measurement part 702; frequency analysis part 703; electroencephalogram evaluation part 704; scaling width change part A; basic information; B: Sound generation rule, C: Sound file.

Claims (19)

Basic information conversion procedure for converting basic information having chaos into data that can be numerically calculated,
Calculating a chaos attractor based on the data converted in the basic information conversion procedure, and generating a chaos property space;
A sound generation method, comprising: a sound generation procedure for generating a sound file from data in the chaos characteristic space generated by the chaos characteristic space generation procedure according to a predetermined sound generation rule. Basic information conversion procedure for converting basic information having fractal properties into numerically operable data,
A fractal characteristic space generation procedure for generating a fractal characteristic space by extracting a fractal characteristic based on the data converted by the basic information conversion procedure;
A sound generation method comprising: a sound generation procedure for generating a sound file from a fractal characteristic space generated by the fractal characteristic space generation procedure according to a predetermined sound generation rule. A biological signal conversion procedure for converting an information provider's individual time-series generated signal into numerically operable data;
Calculating a chaos attractor based on the data converted in the biological signal conversion procedure, and generating a chaos property space;
A sound generation procedure for generating a sound file adapted to the information provider according to a predetermined sound generation rule from data in the chaos characteristic space generated by the chaos characteristic space generation procedure;
A sound generation method comprising: A biological signal conversion procedure for converting an information provider's individual time-series generated signal into numerically operable data;
A fractal characteristic space generation procedure for generating a fractal characteristic space by extracting a self-similar feature based on the data converted by the biological signal conversion procedure;
A sound generation method comprising: a sound generation procedure for generating a sound file adapted to the information provider from a fractal characteristic space generated by the fractal characteristic space generation procedure according to a predetermined sound generation rule. The biological signal conversion procedure includes a biological signal measurement procedure for measuring a biological signal;
A frequency analysis procedure for calculating the biological signal data measured in the biological signal measurement procedure as numerical data for each frequency band;
The sound generation method according to claim 3, further comprising: a sound generation procedure corresponding to a nerve type characteristic in the living body of the individual information provider based on the frequency analysis procedure. The biological signal conversion procedure includes a biological signal measurement procedure for measuring a biological signal;
A frequency analysis procedure for calculating the biological signal data measured in the biological signal measurement procedure as numerical data for each frequency band;
The sound generation method according to claim 4, further comprising: a sound generation procedure corresponding to a nerve type characteristic in the living body of the individual information provider based on the frequency analysis procedure. The chaos characteristic space generation procedure is a state evaluation procedure for comparing the numerical data of nerve type characteristics in the living body of the information provider individual calculated in the frequency analysis procedure, and evaluating the psychosomatic state of the information provider,
The sound generation method according to claim 3, further comprising: a cutting plane changing procedure for changing a plane for cutting the chaotic attractor according to the evaluation in the state evaluation procedure. The fractal characteristic space generation procedure is a state evaluation procedure for comparing the numerical data of nerve type characteristics in the living body of the information provider individual calculated in the frequency analysis procedure, and evaluating the psychosomatic state of the information provider,
The sound generation method according to claim 4, further comprising a scaling width changing procedure for changing a scaling width for extracting a fractal feature in accordance with the evaluation in the state evaluation procedure.   The sound generation procedure includes an interface capable of interacting with an information provider of the biological signal, a condition input procedure that allows a condition relating to sound generation to be input, and the sound generation rule according to the condition input in the condition input procedure A sound generation method according to claim 3, wherein the sound file is generated according to a generation rule set in the generation rule setting procedure.   The sound generation procedure includes an interface capable of interacting with an information provider of the biological signal, a condition input procedure that allows a condition relating to sound generation to be input, and the sound generation rule according to the condition input in the condition input procedure The sound generation method according to claim 4, further comprising: a generation rule setting procedure for setting the sound file, wherein the sound file is generated according to the generation rule set in the generation rule setting procedure.   The sound generation method according to claim 7, wherein in the state evaluation procedure, the psychosomatic state is evaluated based on an appearance rate of an α wave of an electroencephalogram.   The sound generation method according to claim 8, wherein in the state evaluation procedure, a psychosomatic state is evaluated based on an appearance rate of an α wave of an electroencephalogram.   The sound generation method according to any one of claims 3 to 10, wherein at least one of a pulse wave, an electrocardiogram, an electroencephalogram, an electromyogram, and respiration is used as the biological signal.   A computer-readable storage medium having recorded thereon a program for causing a computer to execute at least one of the sound generation methods according to any one of claims 1 to 12.   14. A computer-readable storage medium on which a program for causing a computer to execute the sound generation method according to claim 13 is recorded.   A means for measuring a biological signal, a computer for executing at least one of the sound generation methods according to any one of claims 1 to 12, and a means for reproducing the sound generated by the sound generation method A stand-alone sound generating / reproducing apparatus comprising: a means for measuring an individual of an information provider who provides the biological signal and listens to the sound.   14. A means for measuring a biological signal, a computer for executing the sound generation method according to claim 13, a means for reproducing the sound generated by the sound generation method, and information for providing the biological signal and listening to the sound A stand-alone type sound generating and reproducing apparatus comprising: means for measuring a provider's individual state.   A server computer that executes at least one of the sound generation methods according to any one of claims 1 to 12, a means for measuring a biological signal necessary for the sound generation method, and a means for reproducing sound. And a means for executing on a remote computer connected by a computer network.   A server computer for executing the sound generation method according to claim 13, and a remote computer connected by a computer network comprising means for measuring a biological signal necessary for the sound generation method and means for reproducing the sound. And a network distribution type sound generation / reproduction system.

Images