WO2019003418A1

WO2019003418A1 - Sound information processing device, sound information processing method, and program

Info

Publication number: WO2019003418A1
Application number: PCT/JP2017/024150
Authority: WO
Inventors: 裕子石若
Original assignee: ソフトバンク株式会社
Priority date: 2017-06-30
Filing date: 2017-06-30
Publication date: 2019-01-03
Also published as: JP6912569B2; JPWO2019003418A1

Abstract

[Problem] Until now, sound information which is obtained by imitating the functions of the ears or the brain of a human could not be obtained. [Solution] Sound information obtained by imitating the functions of the ears or the brain of a human can be obtained by means of a sound information processing device provided with: a reception unit which sequentially receives sound information; a modulation unit which divides the sound information received by a reception unit 2 into at least two frequency bands, acquires at least two pieces of partial sound information, synthesizes at least two pieces of partial sound information having different frequency bands and generated at different times, and acquires output sound information; and an output unit which outputs the output sound information.

Description

Sound information processing apparatus, sound information processing method, and program

The present invention relates to a sound information processing apparatus or the like that processes and outputs received sound information.

Conventionally, the brain activity can be maintained while enjoying for a long time by activating the entire brain in a well-balanced manner in a balanced manner, without discarding the original sound information, and being able to select the load according to the level of the listener. There have been audio signal processing methods for video coding.

Such a technology is an audio signal processing method of dividing an audio signal at a predetermined time, dividing it into right and left, and shifting the audio signal distributed left and right thereafter in time, and continuously changing the overlapping rate from 50% to 100%. I assume. The audio signal processing apparatus according to this embodiment is an input unit for inputting headphone output such as a CD player, a storage unit for storing the input signal, a processing unit for dividing and assembling the stored audio signal into predetermined time, and removing noise It has an output unit for outputting the voice signal, an operation display unit for setting the operation of the apparatus and displaying the state, and a control unit for monitoring and controlling the operation of each of these parts (see Patent Document 1).

JP 2008-116887 A

However, in the prior art, it has not been possible to obtain sound information obtained by imitating the function of the human ear or brain.

The sound information processing apparatus according to the first aspect of the present invention sequentially receives the sound information, the receiving unit, and divides the sound information received by the receiving unit into two or more frequency bands, and acquires two or more partial sound information. The sound information processing apparatus includes a changing unit that synthesizes two or more partial sound information having different frequency bands at different timings and acquires output sound information, and an output unit that outputs the output sound information.

With this configuration, it is possible to obtain sound information obtained by imitating the function of the human ear or brain.

In the sound information processing apparatus according to the second aspect of the present invention, the changing unit divides the sound information received by the receiving unit into two or more frequency bands and the two or more partial sound information according to the first invention. Output sound information by combining partial sound information of high frequency band among sound information acquired and received earlier in time and partial sound information of low frequency band among sound information received later in time Is a sound information processing apparatus that acquires

In the sound information processing apparatus according to the third aspect of the present invention, in the first or second aspect, the changing unit divides the sound information received by the receiving unit into two or more frequency bands. First division means for acquiring partial sound information and temporarily accumulating the first cue, and temporally dividing the sound information received by the reception unit into two or more, and acquiring two or more second partial sound information; The second division means temporarily accumulated in the second cue, one or more first partial sound information of the first cue, and one or more second partial sound information of the second cue are sequentially extracted, synthesized, and output A sound information processing apparatus comprising: synthesis means for acquiring sound information.

In the sound information processing apparatus according to the fourth aspect of the present invention, the sound information processing apparatus according to any one of the first to third aspects has a soma identifier for identifying soma and firing condition information on a condition for firing the soma. It has a soma related information storage unit in which two or more soma related information are stored, a combination information storage unit in which one or more combination information specifying a combination between two or more soma is stored, and one or more soma identifiers The output unit further includes an output management information storage unit storing one or more output management information having an output condition which is a condition for output using a firing pattern and output information which is output information, and the change unit The sound information received by the reception unit is divided into two or more frequency bands, and the first division unit for acquiring two or more first partial sound information and the sound information received by the reception unit are temporally divided into two or more , The second division means for acquiring the above-mentioned second partial sound information, the characteristic information acquisition means for acquiring one or more characteristic information from each partial sound information which is the first partial sound information or the second partial sound information, and the partial sound For each piece of information, obtain one or more soma identifiers that identify the soma to be fired first, and one or more obtained from one or more feature information passed from one or more other soma or one or more feature information Information transmission means for acquiring the feature information of one or more soma identifiers of the one or more soma to be judged for firing, and one or more feature information acquired by the information transmission means for each of the two or more pieces of feature information; Judgment means for judging whether or not the soma identified by each soma identifier is ignited using the one or more soma identifiers acquired by the information transmission means and the paired firing condition information, and for each partial sound information , A firing pattern acquiring unit for acquiring a firing pattern including one or more soma identifiers for identifying a soma determined by the breaking unit to fire, output information corresponding to the firing pattern acquired by the firing pattern acquiring unit for each partial sound information It is a sound information processing apparatus comprising: synthesis means for acquiring output sound information from the output management information storage unit and acquiring the acquired one or more output information.

According to the sound information processing apparatus according to the present invention, sound information obtained by imitating the function of the human ear or brain can be obtained.

Block diagram of sound information processing apparatus A according to the first embodiment A flowchart for explaining the first process of the same sound information processing apparatus A A flowchart describing the second process of the sound information processing apparatus A in the same manner. Block diagram of sound information processing apparatus B in the second embodiment Block diagram of the processing unit 6 Flow chart for explaining the operation of the same sound information processing apparatus B A flowchart describing details of an example of the same ignition transmission process A flowchart explaining the details of the same fire determination processing A flowchart describing details of an example of the same firing pattern process A flowchart describing an example of the same soma growth process A flowchart explaining an example of the same AXON growth processing A flowchart describing an example of the same Dendrites growth process A flowchart describing an example of the same soma combining process Flow chart explaining an example of same glial cell growth processing Flow chart explaining an example of the same apoptosis processing Overview of computer system in the above embodiment Block diagram of the same computer system

Hereinafter, embodiments of the sound information processing apparatus and the like will be described with reference to the drawings. In addition, since the component which attached the same code in embodiment performs the same operation | movement, description for the second time may be abbreviate | omitted.

Embodiment 1

In the present embodiment, a sound information processing apparatus will be described which divides received sound information into two or more frequency bands, shifts them in time, combines the shifted sounds, and sequentially outputs them. In the present sound information processing apparatus, normally, the partial sound information of the high frequency band received earlier in time and the partial sound information of the low frequency band received later in time are synthesized.

Further, in the present embodiment, there are a buffer in which partial sound information with high time resolution is stored and a buffer in which partial sound information with high frequency resolution is stored, and partial sound information of each buffer is acquired and synthesized. The sound information processing apparatus will be described.

FIG. 1 is a block diagram of a sound information processing apparatus A according to the present embodiment. The sound information processing apparatus A includes a storage unit 1, a reception unit 2, a change unit 3, and an output unit 4. The changing unit 3 includes a first dividing unit 31, a second dividing unit 32, and a combining unit 33. The changing unit 3 may have three or more dividing units including not only the first dividing unit 31 and the second dividing unit 32 but also a third dividing unit and the like. An example of processing in the case of having three or more dividing means will be described later.

The storage unit 1 stores various types of information. The various types of information are, for example, sound information received by the reception unit 2 and partial sound information to be described later. The storage unit 1 has two or more queues described later.

The receiving unit 2 sequentially receives sound information. The receiving unit 2 may be considered to receive, for example, a stream of sound information. For example, the receiving unit 2 may be considered to receive sound information on which FFT is applied to an electronic signal of analog sound. That is, the sound information received by the receiving unit 2 may be an analog electrical signal or digital data.

Here, acceptance means, for example, acceptance of information input from an input device such as a microphone, reception of information transmitted via a wired or wireless communication line, or from a recording medium such as an optical disc, a magnetic disc, or a semiconductor memory. It is a concept including reception of read information and the like.

The input means may be anything such as a microphone or a menu screen. The reception unit 2 can be realized by a device driver of input means such as a microphone, control software of a menu screen, or the like.

The changing unit 3 divides the sound information received by the receiving unit 2 into two or more frequency bands, and acquires two or more partial sound information. Next, the changing unit 3 synthesizes two or more partial sound information having different frequency bands at different timings to obtain output sound information. The output sound information is sound information to be output. Note that the changing unit 3 may acquire two or more partial sound information from the sound information of digital data, or apply, for example, an FFT to the sound information of an analog electrical signal, and the information obtained by the FFT It may be divided into two or more frequency bands, and two or more partial sound information may be acquired. The partial sound information is digital data acquired from the sound information.

The changing unit 3 normally divides the sound information received by the receiving unit 2 into two or more frequency bands, acquires two or more partial sound information, and high frequency band of the sound information received earlier in time The partial sound information of is synthesized with the partial sound information of the low frequency band of the sound information received later in time to obtain the output sound information. In addition, such processing applies a property in which the sound information in the frequency band is processed earlier than the sound information in the low frequency in the ear.

For example, the changing unit 3 applies a plurality of STFTs (short time Fourier transform) in parallel to the sound information received by the receiving unit 2. Here, for example, the changing unit 3 applies (1) STFT with high time resolution to the sound information received by the receiving unit 2, (2) applies STFT with high frequency resolution, (3) (1) It is assumed that STFT is applied to the sound information in parallel three times in a manner such as to apply STFT in the middle between (2) and (2). In (1), (2) and (3), the rounding widths are different. For example, in the case of (3), the change unit 3 treats + -6% before and after as “440 hz” centering on “frequency = 440 hz”. The processing applied to the STFT is processing of division means such as a first division means 31 described later and a second division means 32 described later.

After applying the STFT, the changing unit 3 adds to the next input data, using, for example, the delay of each frequency and the echo parameter (liquid specific). The addition is a process of the combining unit 33 described later.

The processing result of (1) may be the Octopus feature amount, the processing result of (2) may be the Stellate feature amount, and the processing result of (3) may be the Bushy feature amount. Octopus, Stellate and Bushy are cells which respectively constitute a cochlear nucleus (Cochlear nucleus).

The first dividing unit 31 constituting the changing unit 3 divides the sound information received by the receiving unit 2 into two or more frequency bands, acquires two or more pieces of first partial sound information, and temporarily stores them in the first queue. . When there are a plurality of first cues, the first dividing means 31 accumulates first partial sound information of two or more different frequency bands in the cue corresponding to each frequency band. The queue is determined by the frequency band. The first dividing means 31 may apply STFT having a high frequency resolution in (2) above.

The second dividing means 32 temporally divides the sound information received by the receiving unit 2 into two or more, acquires two or more pieces of second partial sound information, and temporarily accumulates them in a second queue. When there are a plurality of second cues, the second dividing means 32 accumulates second partial sound information of two or more different frequency bands in the corresponding cues. The second dividing means 32 may apply STFT having a high time resolution in (1) above.

The division of sound information by the second division means 32 is higher in time resolution than the division of sound information by the first division means 31. Further, the division of the sound information by the first division means 31 has higher frequency resolution than the division of the sound information by the second division means 32.

The synthesis means 33 sequentially extracts and synthesizes the first partial sound information of one or more first cues and the second partial sound information of one or more second cues, and outputs the output sound information. get. The first partial sound information and the second partial sound information are collectively referred to as partial sound information as appropriate. Also, the synthesis of two or more partial sound information is usually the addition of two or more partial sound information. However, the synthesis of two or more partial sound information may be, for example, a process of adding two or more partial sound information with a weight.

The synthesizing unit 33 sequentially extracts, for example, the first partial sound information of each of the two or more first cues and the second partial sound information of the second cue, synthesizes them, and acquires output sound information. The two or more pieces of first partial sound information of the two or more first cues used for synthesis by the synthesizing means 33 are the first partial sound information of the high frequency band of the sound information received earlier in time; It is partial sound information of a low frequency band among sound information received later in time.

It is preferable that the combining means 33 takes out from the first cue at a faster pace as the first partial sound information of the high frequency band as compared with the partial sound information of the low frequency band and uses it for synthesis.

The output unit 4 outputs the output sound information acquired by the synthesizing unit 33. Here, the output is a concept including delivery of the processing result to another processing device or another program, sound output, transmission to an external device, storage in a recording medium, and the like. The processing performed by another processing device or another program is, for example, the realization of the function of the brain described in the second embodiment. That is, delivery of the processing result is, for example, delivery of the processing result to a processing apparatus or program using the soma group described in the second embodiment.

The storage unit 1 is preferably a non-volatile storage medium, but can be realized also as a volatile storage medium. The process of storing information in the storage unit 1 does not matter. For example, information may be stored in the storage unit 1 via a recording medium, or information transmitted via a communication line may be stored in the storage unit 1, or The information input via the input device may be stored in the storage unit 1.

The changing unit 3, the first dividing unit 31, the second dividing unit 32, and the combining unit 33 can be usually realized by an MPU, a memory or the like. Typically, the processing procedure of the change unit 3 or the like is realized by software, and the software is stored in a storage medium such as a ROM. However, it may be realized by hardware (a dedicated circuit).

The output unit 4 may be realized by an MPU, a memory or the like, or may be realized by driver software of an output device such as a speaker or driver software of an output device and an output device or the like.

Next, an operation of the sound information processing apparatus A receiving sound information and accumulating partial sound information in a cue will be described with reference to the flowchart of FIG.

(Step S201) The receiving unit 2 determines whether sound information has been received. If the sound information is received, the process goes to step S202. If the sound information is not received, the process returns to step S201.

(Step S202) The first dividing unit 31 divides the sound information received by the receiving unit 2 into sound information of two or more frequency bands, and acquires two or more pieces of first partial sound information corresponding to each frequency band.

(Step S203) The first dividing unit 31 temporarily accumulates the two or more pieces of first partial sound information acquired in step S202 in one or more first cues. The first division means 31 temporarily accumulates in the cue so that the first partial sound information in the high frequency frequency band is taken out earlier in time as compared to the first partial sound information in the low frequency frequency band. . For example, the first division means 31 temporarily accumulates the first partial sound information of the high frequency frequency band in a fast cue of the information output rotation, and the first partial sound information of the low frequency frequency band of the information output rotation Temporarily accumulate in a slow queue. Note that a fast rotating queue is a queue with a short cycle of retrieving information. Also, a slow rotation queue is a queue that has a long information retrieval cycle.

(Step S204) The second dividing unit 32 divides the sound information received by the receiving unit 2 into two or more time zones, and acquires two or more pieces of second partial sound information. The second division means 32 may divide each piece of second partial sound information of each time zone into information of two or more frequency bands, and may obtain further divided second partial sound information.

(Step S205) The second division means 32 temporarily accumulates the two or more pieces of second partial sound information acquired in step S204 in one or more second cues. It returns to step S201. Note that the second division means 32 temporarily sets in the queue the second partial sound information in the high frequency frequency band to be taken out earlier in time as compared to the second partial sound information in the low frequency frequency band. You may accumulate.

In the flowchart of FIG. 2, the processing is terminated by an interruption of power off or processing termination.

Next, an operation of the sound information processing apparatus A synthesizing partial sound information of a plurality of cues, constructing output sound information, and outputting the sound information will be described with reference to the flowchart of FIG.

(Step S301) The combining unit 33 determines whether or not partial sound information exists in any of two or more cues. If partial sound information exists, the process goes to step S302, and if partial sound information does not exist, the process returns to step S301.

(Step S302) The combining unit 33 substitutes 1 into the counter i.

(Step S303) The combining unit 33 determines whether the i-th cue storing partial sound information used for combining the output sound information is present. If the i-th queue exists, the process goes to step S304, and if the i-th queue does not exist, the process goes to step S306.

(Step S304) The synthesizing unit 33 acquires partial sound information from the i-th cue. Here, it is preferable not to acquire partial sound information from the cue every time this loop from a cue of slow rotation of information output (usually a cue corresponding to a low frequency band). It goes without saying that the acquired partial sound information is deleted from the cue.

(Step S305) The combining unit 33 increments the counter i by one. It returns to step S303.

(Step S306) The combining unit 33 combines the two or more partial sound information acquired in step S304, and acquires output sound information.

(Step S307) The output unit 4 outputs the output sound information acquired in step S306. It returns to step S301.

In the flowchart of FIG. 3, the processing is ended by an interruption of power off or processing end.

Further, the operation of the flowchart of FIG. 2 and the operation of the flowchart of FIG. 3 are usually executed in parallel.

As described above, according to the present embodiment, it is possible to obtain sound information obtained by imitating the function of the human ear or brain.

In addition, when the output unit 4 passes output sound information to another processing device or another program, the processing performed by the other processing device or the other program on the output sound information does not matter. . Other processing devices and other programs perform in-brain processing using, for example, the soma group described in the second embodiment.

The processing in the present embodiment may be realized by software. Then, this software may be distributed by software download or the like. Also, the software may be distributed by being recorded on a recording medium such as a CD-ROM. This applies to the other embodiments in the present specification. The software for realizing the sound information processing apparatus A in the present embodiment is a program as follows. That is, this program divides the computer into a reception unit that receives sound information sequentially, and divides the sound information received by the reception unit into two or more frequency bands, and acquires two or more partial sound information The program is configured to function as a changing unit that synthesizes two or more partial sound information having different frequency bands at different timings to obtain output sound information, and an output unit that outputs the output sound information.

Further, in the program, the changing unit divides the sound information received by the receiving unit into two or more frequency bands, acquires two or more partial sound information, and among the sound information received earlier in time A program that causes a computer to function as synthesizing output of partial sound information in the high frequency band and partial sound information in the low frequency band of sound information received later in time to obtain output sound information Is preferred.

Further, in the above program, the changing unit divides the sound information received by the receiving unit into two or more frequency bands, acquires two or more first partial sound information, and temporarily stores the first partial sound information in a first queue. A division unit, a second division unit for temporally dividing the sound information received by the reception unit into two or more, acquiring two or more second partial sound information, and temporarily storing the second partial sound information in the second queue; As synthesizing means for sequentially extracting and combining one or more first partial sound information of one cue and one or more second partial sound information of the second cue, and obtaining output sound information, It is preferable that the program is a program that causes a computer to function.

Second Embodiment

The present embodiment differs from the first embodiment in that a structure or function in the brain is used. In the present embodiment, the function to be realized is the same as that of the first embodiment. That is, also in the present embodiment, a sound information processing apparatus will be described which divides the received sound information into two or more frequency bands, shifts them in time, combines the shifted sounds, and sequentially outputs them.

The structures, functions, and the like in the brain described in the present embodiment are as follows. That is, in the present embodiment, the firing conditions of one or more soma (may be referred to as a nerve cell body) are stored, and each soma fires using one or more information obtained from the input. A sound information processing apparatus that determines output information by using the firing pattern, determines output information from the one or more output information, and outputs the output information. explain.

Further, in the present embodiment, a sound information processing apparatus will be described in which the process of transmitting feature information is performed using ignition start point information that manages information of soma to be fired first corresponding to the feature information.

Further, in the present embodiment, a sound information processing apparatus that performs processing of calculating two or more pieces of feature information transmitted to soma via two or more links in soma will be described.

Further, in the present embodiment, a sound information processing apparatus will be described in which there are two or more soma groups (may be referred to as nerve cell groups) and information is passed between the coupled soma groups.

Moreover, in the present embodiment, the information transfer is performed by the link having AXON (may be referred to as axon) and Dentrites (may be referred to as dendrite). The sound information processing apparatus to be used will be described.

Further, in the present embodiment, a sound information processing apparatus that performs information transmission in consideration of synapses and spines will be described.

Further, in the present embodiment, a sound information processing apparatus will be described in which the operated element does not operate if time has not elapsed so as to satisfy the predetermined condition. The elements are, for example, soma, AXON, Dendrites.

Also, in the present embodiment, a sound information processing apparatus will be described that raises the firing probability of soma that has been fired.

Further, in the present embodiment, a sound information processing apparatus will be described which outputs different output sound information based on the firing pattern and one or more pieces of external information. The external information is, for example, weather, temperature, scenery of the outside world, smell, sound, light and the like.

Further, in the present embodiment, a sound information processing apparatus having a learning function will be described.

Further, in the present embodiment, a sound information processing apparatus in which elements are automatically grown will be described. The trigger for growth is firing or receiving information.

Further, in the present embodiment, a sound information processing apparatus which has glial cell information and whose glial cell information affects the growth will be described.

Further, in the present embodiment, an element has position information, and a sound information processing apparatus that realizes the growth and combination of elements using position information will be described.

Also, in the present embodiment, a sound information processing apparatus that imitates the actual growth method such as AXON will be described.

Furthermore, in the present embodiment, a sound information processing apparatus that mimics the apoptosis treatment of cells such as soma will be described. Here, for example, soma disappears when there is a large number of soma so as to satisfy a predetermined condition, or soma which has a small number of times of firing so as to satisfy a predetermined condition so as not to be connected to Dendrites Disappear, or the connected AXON disappears a soma that has not reached a predetermined goal.

FIG. 4 is a block diagram of the sound information processing apparatus B in the present embodiment. FIG. 5 is a block diagram of the processing unit 6.

The sound information processing apparatus B includes a storage unit 5, a reception unit 2, a processing unit 6, and an output unit 4.

The storage unit 5 includes a soma related information storage unit 51, a soma group information storage unit 52, a combined information storage unit 53, a glial cell information storage unit 54, an ignition start point information storage unit 55, an output management information storage unit 56, and a learning condition storage unit. 57, a learning information storage unit 58, a firing information storage unit 59, and a combined use information storage unit 60.

The receiving unit 2 includes an input information receiving unit 21.

The processing unit 6 includes a change unit 61, a learning detection unit 62, a learning information storage unit 63, a growth unit 64, an apoptosis processing unit 65, and a firing information storage unit 66.

The change unit 61 includes a first division unit 31, a second division unit 32, a feature information acquisition unit 611, an information transmission unit 612, a soma calculation unit 613, a determination unit 614, a firing probability change unit 615, a firing pattern acquisition unit 616, and a combination. A unit 617 and a control unit 618 are provided.

The information transfer means 612 includes an ignition start point soma determination means 6121, a connection detection means 6122, and a transfer information acquisition means 6123.

The storage unit 5 stores various types of information. The various information includes, for example, sound information received by the reception unit 2, partial sound information, soma related information to be described later, soma group information to be described later, coupling information to be described later, glial cell information to be described later, ignition starting point information to be described later Output management information to be output, learning conditions to be described later, learning information to be described later, firing information to be described later, usage combination information to be described later. The storage unit 5 usually has two or more queues.

The soma related information storage unit 51 stores two or more soma related information. Here, the soma related information is information on soma. The soma related information has a soma identifier and firing condition information. The soma related information usually includes one or more Dendrites information and one or more AXON information. Dendrites information is information on Dendrites that realizes the input of information to soma. Dendrites information has Dendrites identifier. The Dendrites identifier is information for identifying Dendrites, and is, for example, an ID, a name, or the like. Preferably, the Dendrites information comprises Dendrites location information. AXON information is information related to AXON that realizes output of information from soma. The AXON information has an AXON identifier. The AXON identifier is information for identifying AXON, and is, for example, an ID, a name or the like. The AXON information preferably includes AXON position information. The soma related information may have goal information. Goal information is information for specifying a goal. The goal is the destination of AXON or Dentrites connected to soma. Goal information is information indicating a position. The goal information is, for example, position information. The goal information is, for example, three-dimensional coordinate values (x, y, z) or two-dimensional coordinate values (x, y) or four-dimensional quaternions (x, y, x, w).

In addition, when expressing the connection between soma by the information of one link, the soma related information may not have Dentrites information or AXON information.

In addition, the soma related information may have a synapse identifier that identifies a synapse or a spine identifier that identifies a spine in association with AXON information or Dendrites information. Usually, synapse identifiers correspond to AXON information. Also, usually, spine identifiers correspond to Dendrites information.

Also, the soma related information may have a soma group identifier that identifies the soma group to which the soma belongs. Also, the soma related information may be associated with the soma group identifier.

Moreover, it is preferable that the soma related information has soma position information indicating the position of soma. The soma position information is, for example, three-dimensional coordinate values (x, y, z) or two-dimensional coordinate values (x, y) or four-dimensional quaternions (x, y, x, w).

Also, Dendrites position information is information for specifying the position of Dentrites, and for example, one or more three-dimensional coordinate values (x, y, z), or one or more two-dimensional coordinate values (x) , Y). When Dendrites position information has two or more coordinate values, Dendrites is a line connecting points of the two or more coordinate values.

Further, AXON position information is information for specifying the position of AXON, and, for example, one or more three-dimensional coordinate values (x, y, z), or one or more two-dimensional coordinate values (x) , Y). When AXON position information has two or more coordinate values, AXON is a line connecting points of two or more coordinate values.

Note that Dendrites and AXON may be branched. Each position information when Dendrites and AXON branch may be represented by three or four or more coordinate values. However, it does not matter how to express Dendrites position information and AXON position information.

Also, the soma identifier is information that identifies soma. The soma identifier is, for example, an ID, a name or the like. The firing condition information is information on the condition for firing the soma. The firing condition information usually has one or more pieces of feature information. The characteristic information may be information having an information identifier for identifying the information and an amount of information indicating the size of the information, or may be information of only the amount of information indicating the size of the information. The amount of information is, for example, a numerical value greater than zero. The firing condition information is, for example, “feature information> = 0.5”, “feature information> 0.7”, “information amount> = 0.5”, “information amount> 0.7” “(information identifier = A & information amount > = 0.5) & (information identifier = B & amount of information> 0.8) etc. The feature information that constitutes the firing condition information is the amount of information. The feature information is, for example, a feature amount, but may be input information itself. The feature amount is, for example, a feature amount of sound as a result of sound analysis, and a feature amount of an image as a result of image analysis. Needless to say, the sound may be voice. The firing condition information preferably includes firing probability information. The firing probability information is information on the probability of firing. The firing probability information may be the firing probability itself or a value obtained by converting the firing probability by a function or the like. With reference to the firing probability information, it is preferable that the soma fires or does not fire even if the feature information is the same with the probability indicated by the firing probability information.

Two or more pieces of soma related information may be grouped. "Grouped" is, for example, that the soma related information is associated with the soma group identifier. Correspondence is good meaning as long as correspondence is possible. The soma group identifier is information identifying the soma group to which the soma belongs. "Grouped" means, for example, that goal information possessed by soma related information is identical. Also, "grouped" means, for example, having the same soma group identifier, and "grouped" means, for example, being linked by a link. Needless to say, there is no limitation on the method and data structure of grouping of soma related information. The soma related information may have a soma group identifier.

In addition, it is preferable that the soma related information have stored energy amount information indicating the amount of energy held by the soma. In addition, it is preferable that the soma related information have necessary energy amount information indicating an energy amount necessary for ignition. In addition, it is preferable that the Dendrites information includes possessed energy amount information indicating the amount of energy held by Dendrites. In addition, it is preferable that Dendrites information has necessary energy amount information necessary for performing information transmission using Dendrites. Further, it is preferable that the AXON information includes possessed energy amount information indicating the amount of energy held by AXON. Furthermore, it is preferable that AXON information has necessary energy amount information necessary for performing information transmission using AXON.

The soma group information storage unit 52 stores two or more pieces of soma group information. The soma group information includes soma group identifiers and goal information. The goal information is information for specifying the extension destination of AXON or Dendrites connected to the soma belonging to the soma group. The goal information is, for example, one or more three-dimensional coordinate values (x, y, z), or one or more two-dimensional coordinate values (x, y). The goal information may be, for example, information indicating a direction.

The combined information storage unit 53 stores one or more pieces of combined information. Binding information is information that identifies binding between two or more soma. The binding information may be information specifying binding between one soma AXON and another soma Dendrites. Such information is also information specifying the binding between soma. The binding information may be information specifying the binding between one synapse and one spine. Such information is also information specifying the binding between soma. The binding information has, for example, two soma identifiers to be bound. Further, the connection information includes, for example, an AXON identifier of AXON and a Dendrites identifier of Dendrites to be associated with the AXON. Further, the connection information includes, for example, a synapse identifier of a synapse, and a spine identifier of a spine capable of transmitting information between the synapse. The combined information may comprise information transfer probability information. The information transfer probability information is information on the probability of performing information transfer between one soma and another soma. The information transfer probability information may be information on the probability of performing information transfer between AXON and Dendrites. Also in this case, the information transfer probability information is information on the probability of performing information transfer between one soma and another soma. Further, the information transfer probability information may be information on the probability of performing information transfer between the synapse and the spine. Also in this case, the information transfer probability information is information on the probability of performing information transfer between one soma and another soma. Here, the bonding direction between soma is usually one direction.

The connection information may be information indicating the connection between soma and AXON. In such a case, the combined information includes the soma identifier and the AXON identifier. The connection information may also be information indicating the connection between soma and Dendrites. In such a case, the binding information includes the soma identifier and the Dendrites identifier.

The binding information may be information identifying the binding between glial cells and AXON or Dendrites. In such a case, the binding information includes, for example, a glial cell identifier that identifies glial cell information and an AXON identifier. The binding information may include, for example, a glial cell identifier and a Dendrites identifier.

The glial cell information storage unit 54 stores one or more glial cell information. Glial cell information is information on glial cells. It is preferred that the glial cell information has a glial cell identifier that identifies glial cells. The glial cell information has, for example, a soma identifier that identifies soma to bind, or a binding information identifier that identifies binding information to bind. The glial cell information has, for example, an AXON identifier of AXON that binds to the glial cells, or a Dendrites identifier of Dendrites that binds to the glial cells. Also, it is preferable that glial cell information has a glial cell type identifier that identifies the glial cell type. The types of glial cells are, for example, oligodendrocites (hereinafter referred to as "oligo" as appropriate) and astrocites. Oligo is a cell that can be connected to axon. Astrocites are cells that can connect to soma or Dendrites. Also, it is preferable that glial cell information has glial cell position information that specifies the position of glial cells. In particular, it is preferable that oligo glial cell information has glial cell position information. In addition, glial cell information may have length information indicating the length of one or more hands. In addition, glial cell information may include information indicating the number of hands leaving glial cells. Then, in general, it is preferable that the glial cells do not grow when the total length of the glial cells calculated from the hand length information reaches the threshold.

The firing start point information storage unit 55 stores one or more pieces of firing start point information. The firing start point information has an information identifier for identifying feature information, and one or more soma identifiers for identifying soma to be fired when the feature information is accepted. The information identifier is, for example, information for specifying the type of feature amount of sound information, and is, for example, information for identifying sound information of a specific frequency band.

The output management information storage unit 56 stores one or more output management information. The output management information is information having an output condition and output information. The output management information may be information on a pair of output conditions and output information.

The output condition is a condition used to determine output information. The output condition is a condition for output using the firing pattern. The output condition may be a firing pattern itself or information having a firing pattern and output probability information. The output probability information is information on the probability for acquiring the output information. The output condition may be information on the lower limit of the firing pattern and the number of soma identifiers included in the firing pattern to be applied, information on the lower limit of the ratio of the soma identifier in the firing pattern to the firing pattern to be applied, or the like. The firing pattern has one or more soma identifiers. The firing pattern is a pattern of firing of one or more soma.

The output condition may be a condition using a firing pattern and information on one or more pieces of external information. External information is external information. External information may be referred to as user context. The external information is, for example, temperature, weather, smell and the like. The external information is usually information that can be accepted other than the input information when the input information is accepted.

The output information is information corresponding to the firing pattern. The output information is, for example, one or two or more pieces of feature information (may be referred to as a feature amount) of sound information. The output information is, for example, emotion information on human emotion, intention information indicating the intention of the person, action information on the movement of the human body, and the like. Emotion information is, for example, happy, sad, hate, surprise, etc. Emotion information is, for example, an ID for identifying an emotion. The intention information is, for example, information identifying an intention. The action information is, for example, information reflected on the movement of an avatar (character). The technology for operating the avatar is a known technology, and thus the detailed description is omitted.

The learning condition storage unit 57 stores one or more learning conditions. The learning condition is a condition for learning to be performed. The learning condition is a condition using a firing pattern. The learning condition may be the firing pattern itself. The learning condition includes, for example, a firing pattern having one or more soma identifiers, the number of soma requiring firing for learning, or the ratio of soma requiring firing for learning (included in the firing pattern for learning condition) Among the soma identifiers to be sent, the number of fired soma / the number of soma identifiers included in the firing pattern included in the learning condition may be used. Also, the learning conditions may include firing patterns and learning probability information. The learning probability information is information on the probability to decide to learn. If the learning condition has learning probability information, even if the judgment using the firing pattern can be judged as "learning", it is judged probabilistically using the learning probability information and judged as "do not learn" It may be done.

The learning information storage unit 58 stores one or more pieces of learning information. Learning information is learned information. The learning information is information used after learning. The learning information includes input information or one or more pieces of feature information acquired from the input information, and a firing pattern. The feature information here may be an information identifier for identifying the feature information. The characteristic information here may be an information identifier and an information amount. The learning information may have holding time information indicating the holding time of the firing pattern. The holding time is the time to be erased when the firing pattern is not used. Note that, for example, the processing unit 6 performs a process of deleting learning information that has not been used for the time indicated by the holding time information.

The firing information storage unit 59 stores one or more firing information. The firing information here is information on the result of firing. The firing information has a soma identifier that identifies the fired soma. The firing information may typically include timer information indicating when it fires. The timer information may be information indicating a relative time or may be information on a time indicating an absolute time. Note that the firing information in the firing information storage unit 59 may be automatically deleted by the processing unit 6 after a predetermined time has elapsed since the storage.

The usage combination information storage unit 60 stores one or more usage combination information. The use combination information is information indicating a history of use of the combination information for the transmission of information. The use combination information may be information indicating a history of use of AXON or Dentrites for transmission of information. The usage binding information has, for example, a binding information identifier. The usage binding information has, for example, an AXON identifier or / and a Dendrites identifier. The usage binding information has, for example, a synapse identifier or / and a spine identifier. The usage combination information may include, for example, timer information indicating when it was used.

The receiving unit 2 receives various types of information. The various types of information are, for example, input information such as sound information, external information, and the like. The means for inputting various information may be anything such as a microphone, a keyboard, a mouse, a menu screen, or various sensors such as a motion sensor and a temperature sensor. The reception unit 2 can be realized by a device driver of input means such as a microphone or a keyboard, control software of a menu screen, or the like. In addition, reception includes reception of information input from an input device such as a microphone, a keyboard, a mouse, and a touch panel, reception of information transmitted through a wired or wireless communication line, an optical disk, a magnetic disk, a semiconductor memory, etc. It is a concept including reception of information read from a recording medium, and the like.

The input information accepting unit 21 accepts input information. The input information is information to be input. Here, the input information is, for example, sound information. There is no limitation on the data type or data structure of the input information. It is preferable that the input information receiving unit 21 also receive one or more pieces of external information.

The processing unit 6 performs various processes. The various types of processing are, for example, processing performed by the change unit 61, the learning detection unit 62, the learning information storage unit 63, the growth unit 64, the apoptosis processing unit 65, the firing information storage unit 66, and the like.

The changing unit 61 divides the sound information received by the receiving unit 2 into two or more frequency bands, acquires two or more partial sound information, and two or more partial sound information with different frequency bands at different timings. To obtain output sound information.

The changing unit 61 divides the sound information received by the receiving unit 2 into two or more frequency bands, acquires two or more partial sound information, and a portion of the high frequency band of the sound information received earlier in time The sound information and the partial sound information in the low frequency band of the sound information received later in time are synthesized to obtain output sound information.

The feature information acquisition unit 611 acquires one or more pieces of feature information from one or more pieces of partial sound information. The partial sound information is first partial sound information or second partial sound information.

Feature information (which may be referred to as a feature amount) obtained from sound information is, for example, frequency, time, and delay time. In addition, since the process which acquires the feature-value of sound information is a well-known technique, detailed description is abbreviate | omitted.

The information transfer means 612 obtains one or more soma identifiers identifying the soma that fires first. In detail, the information transfer means 612 acquires one or more soma identifiers paired with the information identifier possessed by the feature information acquired by the feature information acquisition means 611 from the firing start point information storage unit 55, and the one or more soma The identifier and the amount of information possessed by the feature information acquired by the feature information acquisition means 611 are paired and acquired. Generally, an amount of information is given to each soma identified by the one or more soma identifiers. Also, the soma identified by the one or more soma identifiers may be an identifier of the soma to be fired first.

The information transfer means 612 is configured to receive one or more pieces of feature information acquired from one or more pieces of feature information or one or more pieces of feature information passed from one or more other soma, and one or more pieces of soma to be subjected to the determination of firing Get the soma identifier.

The information transfer means 612 acquires, for example, one or more feature information acquired by the feature information acquisition means 611 and one or more soma identifiers for identifying a soma to be fired first, for each partial sound information, and one or more One or more pieces of feature information passed from another soma or one or more pieces of feature information acquired from one or more pieces of feature information and a soma identifier of one or more soma to be subjected to the determination of firing are acquired.

The information transfer means 612 includes one or more feature information applied to the soma related information of the soma determined by the determination means 614 to be fired, or one or more feature information acquired from the one or more feature information, and the determination means 614 Then, the determined soma and one or more soma identifiers associated with one or more soma are acquired. The information transfer means 612 acquires the soma identifier of one or more soma connected with the soma judged by the judgment means 614 to fire using the connection information of the connection information storage unit 53.

The information transfer means 612 acquires one feature information acquired by the soma calculation means 613 described later and one or more soma identifiers of one or more soma to be subjected to the determination of the firing. Here, one feature information is usually one information amount.

The information transfer means 612 uses the connection information specifying the connection between the soma group (referred to as the first soma group) including the soma judged to be firing by the determination means 614 and the other soma group (referred to as the second soma group), In some cases, one or more soma soma identifiers included in the second soma group may be obtained. The one or more soma soma identifiers included in the second soma group are identifiers of soma present near the position of the first soma group.

The information transfer means 612 uses the information transfer probability information included in the combined information in the combined information storage unit 53 to probabilistically use the soma identifier of each of one or more soma combined with the soma which the determination means 614 determines to fire. Get to Stochastic acquisition is to use information on probability (here, information transmission probability information) to judge to fire or not to fire. In addition, since the process acquired stochastically is a well-known technique, detailed description is abbreviate | omitted.

Note that, for example, after judgment means 614 to be described later indicates that AXON or Dendrites, or synapse or spine is used history indicating utilization history, the AXON or Dendrites or the synapse or the spine is used recently. It is preferable that the information transfer means 612 perform the process of transferring information to the next soma only when it is determined that the time has not been used for a predetermined time or more or for a time longer than the predetermined time.

In addition, when the information transfer means 612 performs the information transfer process to the next soma, it is preferable that the use combination information be configured and stored in the use combination information storage unit 60. That is, the information transfer means 612 acquires the timer information which shows the present time from the timer which is not shown in figure, when the information transfer process to the following soma is performed. Then, the information transfer means 612 acquires one or more of the used AXON AXON identifier, the used Dendrites Dendrites identifier, the used synapse synapse identifier, or the used spine spine identifier. . Then, the information transfer means 612 configures timer information and usage combination information having the acquired one or more identifiers, and stores the information in the usage combination information storage unit 60. Further, the information transfer means 612 acquires, for example, a combined information identifier of a combination (link) used for information transfer, acquires timer information indicating the current time from a timer not shown, and the combined information identifier and the timer. The usage combination information including the information is configured and stored in the usage combination information storage unit 60.

In addition, when the information transfer means 612 performs information transfer processing to the next soma, the stored energy amount information paired with the AXON identifier of AXON used for the transfer, and Dentrites used for the transfer. It is preferable to reduce the amount of energy indicated by the stored energy amount information paired with the Dendrites identifier of. In addition, it is assumed that the function for reducing the amount of energy is stored in the storage unit 5, for example. Also, the function does not matter. The function is, for example, Equation 1 below.

In Equation 1, E, g and c are parameters, and I is an input signal (request signal). In Equation 1, for example, E = −65 mV, g = 0.025 μS, and C = 0.5 nF.

The firing start point soma determination unit 6121 obtains, from the firing start point information storage unit 55, one or more soma identifiers paired with the information identifiers for identifying the one or more pieces of feature information obtained by the feature information obtaining unit 611. The information identifier may be included in the feature information acquired by the feature information acquisition unit 611, or may be associated with the feature information or the information amount acquired by the feature information acquisition unit 611.

The connection detection unit 6122 detects one or more soma connected to the soma determined by the determination unit 614 to be fired, using the connection information in the connection information storage unit 53. The detection of soma is usually the acquisition of soma identifiers.

The connection detection unit 6122 acquires, for example, the soma identifier of the soma determined by the determination unit 614 to fire, and acquires one or more soma identifiers paired with the soma identifier from the connection information storage unit 53.

The connection detection unit 6122 acquires, for example, the soma identifier of the soma determined by the determination unit 614 to be fired, acquires an AXON identifier that makes a pair with the soma identifier from the soma related information storage unit 51, and makes a pair with the AXON identifier. And the soda identifier paired with the Dendrites identifier is acquired from the soma related information storage unit 51.

The connection detection unit 6122 acquires, for example, the soma identifier of the soma judged by the judgment unit 614 to fire, acquires a synapse identifier that makes a pair with the soma identifier from the soma related information storage unit 51, and makes a pair with the synapse identifier. The spine identifier is acquired from the combined information storage unit 53, and the soma identifier paired with the spine identifier is acquired from the soma related information storage unit 51.

The connection detection unit 6122 acquires, for example, the soma group identifier of the soma group to which the soma determined by the determination unit 614 to be fired belongs. When the soma judged by the judging means 614 to fire is the soma at the end of the soma group (there is no soma transmitting characteristic information in the same soma group), the connection detecting means 6122 is The soma group identifier of the other soma group paired with the soma group identifier identifying the soma group is acquired from the combined information storage unit 53. Next, the binding detection means 6122 may, for example, receive feature information from one or more soma at the end of the soma group identified by the acquired soma group identifier (other soma in the same soma group) Get one or more soma identifiers that identify no soma).

For example, the connection detection unit 6122 stores the soma identifier of the soma determined by the determination unit 614 as firing in the storage unit 1 as the final soma of the information transmission in the soma group to which the soma belongs. The soma group identifier of another soma group paired with the soma group identifier identifying the soma group is acquired from the combined information storage unit 53. Then, for example, among the one or more soma identifiers paired with the acquired soma group identifier of the other soma group, the binding detection means 6122 is a soma that first receives information transmission from the other soma group, When stored in the storage unit 5, the soma identifier of the soma is acquired from the storage unit 5.

Transmission information acquisition means 6123 acquires information used for transmission of information between soma. The transmission information acquisition unit 6123 acquires the feature information to be transmitted and the soma identifier of the transmission destination soma. For example, the transfer information acquisition unit 6123 detects one or more pieces of feature information acquired from one or more pieces of feature information or one or more pieces of feature information applied to the soma related information of the soma determined by the determination unit 614 to fire. The one or more soma identifiers of the one or more soma detected by the means 6122 are acquired.

The soma calculating unit 613 performs an operation on two or more pieces of feature information passed from two or more other pieces of soma, and acquires one piece of feature information. The feature information here is usually the amount of information. That is, the soma calculating unit 613 normally performs an operation on two or more pieces of information passed from two or more other soma, and acquires one piece of information. The operation is a predetermined operation. The operation is, for example, processing of adding two or more information amounts passed from two or more other soma. The operation is, for example, processing of adding two or more information amounts passed from two or more other soma, and then multiplying by a constant less than one.

The determination unit 614 is identified by each soma identifier using the one or more feature information acquired by the information transfer unit 612 and the firing condition information paired with the one or more soma identifiers acquired by the information transfer unit 612. To see if it fires. The one or more pieces of feature information are, for example, the amount of information.

For example, the determination unit 614 acquires, from the soma related information storage unit 51, firing condition information paired with one or more soma identifiers acquired by the information transfer unit 612. Then, the determination unit 614 determines, for example, whether or not the amount of information acquired by the information transfer unit 612 matches the condition indicated by the acquired firing condition information.

It is preferable that the judging means 614 does not judge that an ignition will occur if the time has not elapsed so that the predetermined condition is satisfied, with respect to the soma judged to be an ignition. Not judging that it fires may be judging that it does not fire.

The determination unit 614 refers to, for example, the firing information storage unit 59, and is timer information that is paired with the soma identifier of the determination target soma, and acquires recent timer information. Then, the determination unit 614 obtains, for example, current timer information from a timer (not shown). Then, the determination unit 614 obtains, for example, information of an elapsed time since the latest firing from the current timer information and the latest timer information. Next, the judging means 614 judges whether the elapsed time is less than the threshold or the elapsed time is equal to or less than the threshold, and if it is determined that the elapsed time is less than the threshold or less than the threshold, the soma It is preferable to judge not to fire.

When the firing condition information has firing probability information indicating the probability of firing, the determining unit 614 determines whether or not the soma fires using the firing probability information, so that the determining unit 614 determines one or more of the same. Even if it is determined whether or not the soma fires, it is determined that the soma fires, or it is determined that the soma does not fire, even if the feature information of and the firing condition information are used. That is, even if the same one or more pieces of feature information are given to one soma, the judging means 614 may use the firing probability information corresponding to the soma to judge that it will fire or not. It is suitable.

In addition, it is preferable that the determination unit 614 determines that the apparatus is fired or not according to the external information even if the same one or more pieces of feature information are input.

The determination unit 614 stores, for example, the soma identifier of the fired soma and timer information indicating the time of firing in the storage unit 5 as a pair. That is, when it is determined that the soma has been fired, the determination unit 614 acquires timer information from a timer (not shown). Then, the firing information storage unit 59 stores firing information having the soma identifier of the fired soma and the timer information.

It is preferable that the combined information also does not operate once when it has operated, and if the time has not elapsed enough to satisfy a predetermined condition. That is, the determination means 614 acquires timer information indicating when AXON or Dendrites used for information transmission or synapse or spine was recently used from the combined use information storage unit 60, and indicates timer information indicating the current time. It is preferable to compare and determine that transmission of information using the AXON or the Dendrites or the synapse or the spine is not performed when the time has not elapsed to satisfy the predetermined condition.

Further, it is preferable that the judging means 614 reduce the energy amount indicated by the stored energy amount information included in the soma related information of the soma when it is judged that one soma is ignited. In addition, it is assumed that the function for reducing the amount of energy is stored in the storage unit 5, for example. Also, the function does not matter. The function is, for example, Equation 2 below.

In Equation 2, t is time, and f (t) is the amount of stored energy.

The determination unit 614 uses, for example, one or more pieces of feature information and one or more pieces of soma identifiers and firing condition information to be paired with each of the two or more pieces of feature information to identify the soma identified by each soma identifier. Determine if it will fire. The determination unit 614 determines, for example, whether the one or more feature information acquired by the information transfer unit 612 satisfies the firing condition indicated by the firing condition information paired with the one or more soma identifiers acquired by the information transfer unit 612. To judge.

The firing probability changing unit 615 changes the firing probability information corresponding to the soma determined by the determining unit 614 to fire so that the probability of firing increases. That is, the firing probability changing unit 615 acquires the soma identifier of the soma determined by the determining unit 614 to fire, and changes the firing probability information paired with the soma identifier so that the probability of firing increases. The algorithm for changing the firing probability information is not limited. The firing probability changing means 615 may add a predetermined value to the firing probability information or may add a predetermined ratio value to the firing probability information, or one or more of the accepted features Based on the information, the probability of raising may be acquired. That is, the degree of increase in probability may be constant or may change dynamically.

The firing pattern acquisition unit 616 acquires a firing pattern including one or more soma identifiers for identifying the soma determined by the determination unit 614 to be fired.

The firing pattern acquisition unit 616 applies the input information received by the input information reception unit 21 or one or more feature information acquired from the input information to one or more learning information of the learning information storage unit 58, and the input information or input information It is preferable to obtain a firing pattern corresponding to one or more pieces of feature information obtained from. The firing pattern here is a firing pattern acquired using learning information.

The timing at which the firing pattern acquisition unit 616 acquires the firing pattern does not matter. The firing pattern acquiring unit 616 may periodically acquire the firing pattern, or may irregularly acquire the firing pattern, or acquires the firing pattern every time the determination unit 614 detects the firing of the soma. It is good.

In addition, the temporal width of the firing pattern acquired by the firing pattern acquisition unit 616 does not matter. The firing pattern acquisition unit 616 may obtain, from the firing information storage unit 59, one or more soma identifiers paired with timer information indicating a time within the threshold or a time later than the current time. The firing pattern acquisition unit 616 may obtain one or more soma identifiers included in all the firing information of the firing information storage unit 59 from the firing information storage unit 59. It is preferable that the temporal width of the firing pattern acquired by the firing pattern acquisition unit 616 dynamically change.

For example, the firing pattern acquisition unit 616 acquires, for each partial sound information, a firing pattern including one or more soma identifiers for identifying the soma determined by the determination unit 614 to fire.

The synthesizing unit 617 acquires output information corresponding to the firing pattern acquired by the firing pattern acquiring unit 616 from the output management information storage unit 56 for each partial sound information, and configures output sound information from the acquired one or more output information Do. The output information may be the same type of information as the output sound information, or may be different types of information.

The output sound information corresponding to the firing pattern is usually output sound information that is paired with a firing pattern that is similar to the firing pattern obtained by the firing pattern obtaining unit 616 so as to satisfy a predetermined condition. When the firing pattern A and the firing pattern B are similar so as to satisfy the predetermined condition, for example, soma identifiers whose number is equal to or more than the threshold or more than the threshold among the one or more soma identifiers included in the firing pattern A are fired. This is the case in which the pattern B is included. When the firing pattern A and the firing pattern B are similar so as to satisfy the predetermined condition, for example, soma identifiers whose number is equal to or more than the threshold or more than the threshold among the one or more soma identifiers included in the firing pattern B are fired. In the case of being included in the pattern A. When the firing pattern A and the firing pattern B are similar so as to satisfy the predetermined condition, for example, the one or more soma identifiers included in the firing pattern A have a ratio of at least the threshold or at least a threshold of the soma identifiers. This is the case in which the pattern B is included. When the firing pattern A and the firing pattern B are similar so as to satisfy the predetermined condition, for example, the one or more soma identifiers included in the firing pattern B have a ratio of at least the threshold or at least a threshold of the soma identifiers. In the case of being included in the pattern A.

The combining means 617 detects, for example, a firing pattern similar to the firing pattern obtained by the firing pattern obtaining means 616 so as to satisfy a predetermined condition, and uses output probability information to be paired with the firing pattern to perform probabilistically If it is determined that the output condition is satisfied, output information to be paired with the output condition is acquired from the output management information storage unit 56. Then, the synthesis unit 617 acquires output sound information from the acquired one or more output information. The output sound information may be the same as the output information.

The combining unit 617 determines an output condition that matches the firing pattern acquired by the firing pattern acquisition unit 616 and the one or more external information received by the input information receiving unit 21, and acquires output information that is paired with the output condition. Do. The combining unit 617 determines an output condition from the output conditions stored in the output management information storage unit 56, and acquires output information to be paired with the output condition.

The combining unit 617 detects, from the output management information storage unit 56, for example, a firing pattern similar to the firing pattern acquired by the firing pattern acquisition unit 616 as the firing pattern acquired in advance is one or more to be paired with the firing pattern. When it is determined that the external information of the output information is similar to the external information accepted by the input information accepting unit 21 so as to satisfy the predetermined condition, the firing pattern in the output management information storage unit 56 and the one or more relevant Get output information to be paired with external information of.

Further, the combining unit 617 detects, from the output management information storage unit 56, a firing pattern similar to the firing pattern acquired by the firing pattern acquiring unit 616, for example, so as to satisfy a predetermined condition, and makes a pair with the firing pattern. When it is determined that one or more external information is similar to the one or more external information accepted by the input information accepting unit 21 so as to satisfy a predetermined condition, the firing pattern is paired with the one or more external information Whether or not the output condition is satisfied is determined probabilistically using the output probability information, and when it is determined that the output condition is satisfied, the output information paired with the output condition is output from the output management information storage unit 56 get.

Note that the combining unit 617 can not always obtain output information. In addition, the combining unit 617 sequentially extracts and combines, for example, one or more first partial sound information of the first cue and one or more second partial sound information of the second cue, and acquires output sound information. It is good.

The control unit 618 controls the processing of the determination unit 614, the processing of the firing pattern acquisition unit 616, and the processing of the information transfer unit 612 to be repeated twice or more.

The learning detection unit 62 detects a learning condition in which the firing pattern acquired by the firing pattern acquisition unit 616 matches. Here, the partial soma identifier of the firing pattern acquired by the firing pattern acquisition unit 616 is similar to the state that all or part of the soma identifiers constituting the firing pattern included in the learning condition satisfy predetermined conditions. In this case, the learning detection unit 62 determines that the firing pattern acquired by the firing pattern acquisition unit 616 matches the learning condition. Further, detection of a learning condition is, for example, acquisition of a learning condition identifier that identifies the learning condition, acquisition of information indicating that the learning condition is met, and the like.

When the learning detection unit 62 detects a matching learning condition, the learning information storage unit 63 stores the learning information in the learning information storage unit 58. The learning information includes at least one of input information that is the source of the firing pattern acquired by the firing pattern acquisition unit 616 or one or more feature information acquired from the input information, and at least one of the firing patterns acquired by the firing pattern acquisition unit 616 And a firing pattern having a part soma identifier. The at least one soma identifier is, for example, one or more soma identifiers excluding the one or more soma identifiers used to detect the learning condition from the firing pattern acquired by the firing pattern acquiring unit 616.

The growth unit 64 performs one or more of the soma generation process, the connection information generation process, the connection information growth process, and the glial cell generation process.

The soma generation process is a process of generating soma related information having a soma identifier. The soma generation process generates, for example, a unique soma identifier, and soma position information of a position satisfying a predetermined condition from the position indicated by the soma position information possessed by the soma related information of the division source soma, and the soma concerned This is processing for accumulating soma related information having an identifier and the soma position information in the soma related information storage unit 51.

The soma generation process may be, for example, a process of copying a part of information constituting soma related information of the soma of the division source and generating soma related information of the divided soma having the information. The information to be copied is, for example, firing condition information. In addition, it is preferable that the growing unit 64 sets the soma position information included in the generated soma related information as information of a position not overlapping the position of another element. In addition, it is preferable that the growing unit 64 sets the soma position information included in the generated soma related information to information of a position close enough to satisfy a predetermined condition with the position indicated by the soma position information of the soma at the division source.

The combined information generation process is a process of generating combined information and storing the combined information in the combined information storage unit 53. The combined information generation process is, for example, a process of acquiring soma identifiers of two soma to be combined, generating combined information having the two soma identifiers, and storing the combined information in the combined information storage unit 53. The combined information generation process is, for example, a process of acquiring an AXON identifier of AXON to be combined and a Dendrites identifier of Dendrites, generating combined information having the two identifiers, and storing the combined information in the combined information storage unit 53. In addition, the connection information generation process generates, for example, a synapse identifier of a synapse that transmits information and a spine identifier of a spine to which information is transmitted, generates connection information having the two identifiers, and stores the connection information in the connection information storage unit 53. It is a process.

The combined information growth process is a process of growing combined information. The combined information growing process is, for example, changing the position of AXON position information included in AXON information in a direction in which AXON expands. The combined information growing process is, for example, changing the position of Dendrites position information possessed by Dendrites information in the direction in which Dendrites extends.

The growth part 64 generates soma related information of a divided soma, which is a new soma that has split so many so that the number or frequency of judgment that the judgment means 614 has fired is a predetermined condition, and the soma related information It is preferable to perform soma generation processing to be accumulated in the storage unit 51.

The growth unit 64 generates combination information that combines soma and split soma whose number or frequency of determinations that the determination means 614 has fired is a predetermined condition, and stores the combination information in the combination information storage unit 53. It is preferable to perform information generation processing. The divided soma is a soma from which the soma causing the divided soma is divided.

The growth part 64 generates soma related information of a divided soma which is a new soma obtained by dividing the soma in which the connected glial cell information satisfies a predetermined condition, and accumulates it in the soma related information storage unit 51. It is preferable to perform the generation process.

It is preferable that the growth part 64 performs a combined information growth process to grow AXON or Dendrites which satisfies glial cell information connected thereto in a predetermined condition.

When the growth part 64 is close enough that the AXON position information of one soma and the Dendrites position information of the other soda's Dendrites meet the predetermined condition, the soma identifier of one soma and the soma identifier of another soma It is preferable to perform combined information generation processing for generating combined information including the above and storing the combined information in the combined information storage unit 53.

It is preferable for the growth part 64 to perform the following glial cell generation process. That is, for example, when the amount of energy possessed by an element which is, for example, soma, AXON, or Dentrites decreases to a required energy amount so as to satisfy a predetermined condition, the growing portion 64 connects to the element. Generate cell information. The predetermined condition is, for example, "held energy amount <required energy amount" or "held energy amount <= required energy amount" or "held energy amount-required energy amount <= threshold" or "held energy amount-required Energy amount <threshold. More specifically, for example, the growing unit 64 is a required energy amount possessed by the amount-of-held energy amount information possessed by the information of each element (soma related information, AXON information, or Dendrites information). An identifier (soma identifier, AXON identifier, or Dendrites) for identifying the element if it is determined that the amount of energy required is smaller than the amount of required energy indicated by the amount information, and it is determined that the amount is smaller. Glial cell information having the identifier) is stored in the glial cell information storage unit 54.

The apoptosis processing unit 65 may delete soma related information from the soma related information storage unit 51. When the soma related information is deleted, the apoptosis processing unit 65 performs the soma related information corresponding to the soma related information, the information related to AXON connected to the soma corresponding to the soma related information, and the soma corresponding to the soma related information It is preferable to delete the information about Dendrites connected to. That is, the apoptosis processing unit 65 combines the binding information having the soma identifier included in the deleted soma related information, the binding information including the AXON identifier included in the deleted soma related information, and the binding information including the Dentrites identifier included in the deleted soma related information. The deletion from the combined information storage unit 53 is preferable. When deleting soma related information from the soma related information storage unit 51, the apoptosis processing unit 65 may delete the binding information to AXON and Dendrites connected to the soma corresponding to the soma related information.

In addition, the apoptosis processing unit 65 may delete AXON information. In addition, the apoptosis processing unit 65 may delete Dendrites information. When soma is not apoptotic and only AXON or Dendrites are apoptotic, for example, the apoptosis processing unit 65 deletes the AXON information of the apoptotic AXON or the Dendrites information of the apoptotic Dendrites from the soma related information. In such a case, it is preferable that the apoptosis processing unit 65 delete, for example, the binding information having the AXON identifier of the apoptotic AXON and the binding information having the dentrites identifier of the apoptotic Dendrites from the binding information storage unit 53. .

In addition, it is preferable that the apoptosis processing unit 65 delete, from the glial cell information storage unit 54, information on connection of the glial cells connected to the deleted soma with the soma. That is, for example, the apoptosis processing unit 65 deletes the soma identifier from glial cell information including the soma identifier of the deleted soma. In addition, when AXON or Dendrites is apoptotic, or soma to which AXON or Dendrites is connected is apoptotic, the apoptosis processing unit 65 determines glial cell information including the AXON identifier of the AXON or the Dendrites identifier of the Dendrites. Delete the AXON identifier or Dendrites identifier. That is, when the information of soma or AXON or Dendrites is deleted, it is preferable that the apoptosis processing unit 65 delete the information of connection to soma or AXON or Dendrites from glial cell information.

It is preferable that the apoptosis processing unit 65 delete the soma related information from the soma related information storage unit 51 in accordance with predetermined conditions.

It is preferable that the apoptosis processing unit 65 delete the soma related information from the soma related information storage unit 51 when the soma related information stored in the soma related information storage unit 51 is large as the predetermined condition is satisfied. is there.

It is preferable that the apoptosis processing unit 65 determine a soma not connected to another soma or Dendrites or AXON, and delete the soma related information having the determined soma soma identifier from the soma related information storage unit 51. . The apoptosis processing unit 65 examines, for example, the binding information storage unit 53, acquires a soma identifier whose number of occurrences of the soma identifier is less than or equal to the threshold value, and the soma related information storage unit includes the soma related information including the soma identifier. Remove from 51 In addition, the apoptosis processing unit 65 examines the soma related information storage unit 51, for example, soma related information storage unit having soma related information having only Dendrites information below the threshold or below the threshold, or AXON information below the threshold or below the threshold. Remove from 51

The apoptosis processing unit 65 determines the soma in which the connected AXON has not reached the predetermined goal, and deletes the soma related information having the determined soma soma identifier from the soma related information storage unit 51. Is preferred. For example, the apoptosis processing unit 65 examines the soma related information storage unit 51, compares the AXON position information contained in the soma related information with the goal information, and determines a soma in which AXON has not reached the predetermined goal. The soma related information having the determined soma soma identifier is deleted from the soma related information storage unit 51.

The apoptosis processing unit 65 uses the one or more firing information of the firing information storage unit 59 to determine a soma identifier whose number of times of firing is so small as to satisfy a predetermined condition, and soma related information having the soma identifier is soma. It is preferable to delete from the related information storage unit 51. The apoptosis processing unit 65 determines, for example, a soma identifier whose appearance frequency is less than or equal to the threshold value, and deletes the soma related information having the soma identifier from the soma related information storage unit 51.

The apoptosis processing unit 65 deletes information related to AXON or information related to Dentrites in accordance with predetermined conditions. The information related to AXON is, for example, AXON information, binding information including an AXON identifier, or an AXON identifier in glial cell information. The information regarding Dendrites is, for example, Dendrites information or binding information including Dendrites identifier, Dendrites identifier in glial cell information.

For example, when it is determined that the apoptosis processing unit 65 has become a mitch and it is determined that there is AXON or Dendrites not connected anywhere, the information on the AXON or the information on the Dendrites is deleted. In addition, becoming mitch Michi is that there are multiple phrases so that the number of one or more types of information among soma related information, AXON information, Dendrites information, and glial cell information satisfy a predetermined condition. That is, in the space of the image processing device B (a device imitating a brain), the more the elements satisfy the condition, the free space in the space is less than or less than the threshold. The elements are soma, AXON, Dendrites, glial cells, synapses and spines.

The firing information storage unit 66 constructs firing information having a soma identifier that identifies the soma determined by the determination means 614 to fire, and stores the firing information in the firing information storage unit 59. The firing information storage unit 66 acquires, for example, a soma identifier identifying the soma determined by the determining unit 614 to fire, and acquires timer information indicating the current time from a timer not shown, and the soma identifier and the timer information And stores the firing information in the firing information storage unit 59.

Storage unit 5, soma related information storage unit 51, soma group information storage unit 52, combined information storage unit 53, glial cell information storage unit 54, firing start point information storage unit 55, output management information storage unit 56, learning condition storage unit 57 The learning information storage unit 58, the firing information storage unit 59, and the combined use information storage unit 60 are preferably non-volatile storage media, but can be realized as volatile storage media.

There is no limitation on the process of storing information in the storage unit 5 or the like. For example, the information may be stored in the storage unit 5 or the like through the recording medium, or the information transmitted through the communication line may be stored in the storage unit 5 or the like. Alternatively, the information input via the input device may be stored in the storage unit 5 or the like.

Processing unit 6, change unit 61, learning detection unit 62, learning information storage unit 63, growth unit 64, apoptosis processing unit 65, firing information storage unit 66, feature information acquisition unit 611, information transmission unit 612, soma calculation unit 613, The judging unit 614, the firing probability changing unit 615, the firing pattern acquiring unit 616, the combining unit 617, and the control unit 618 can be usually realized by an MPU, a memory or the like. The processing procedure of the processing unit 6 or the like is usually realized by software, and the software is stored in a storage medium such as a ROM. However, it may be realized by hardware (a dedicated circuit).

Next, the operation of the sound information processing apparatus B will be described using the flowchart of FIG.

(Step S601) The input information accepting unit 21 accepts external information and temporarily accumulates it in the storage unit 5. Here, the input information accepting unit 21 may not accept external information.

(Step S602) The input information receiving unit 21 determines whether input information has been received. If the input information is received, the process goes to step S603, and if the input information is not received, the process goes to step S604. The input information here is, for example, sound information.

(Step S603) The sound information processing apparatus B performs an ignition transmission process. It returns to step S601. The firing transmission process is a process in which information is transmitted between soma. Details of the example of the ignition transmission process will be described with reference to the flowchart of FIG.

(Step S604) The processing unit 6 determines whether or not firing pattern processing is to be performed. If the firing pattern process is to be performed, the process proceeds to step S605, and if the firing pattern process is not to be performed, the process proceeds to step S606. The processing unit 6 may determine that the firing pattern process is always performed, or may determine that the firing pattern process is performed at an interval of a fixed period.

(Step S605) The sound information processing apparatus B performs firing pattern processing. It returns to step S601. The firing pattern process is a process performed using a firing pattern, and includes, for example, a process of determining output information using a firing pattern and outputting the output information, and a learning process. Details of the example of the firing pattern process will be described with reference to the flowchart of FIG.

(Step S606) The processing unit 6 determines whether to perform the growth process and the apoptosis process. When the growth process or the like is performed, the process proceeds to step S607, and when the growth process or the like is not performed, the process returns to step S601. Note that the processing unit 6 may determine to always perform the growth process or the like, or may determine to perform the growth process or the like at regular intervals. There is no limitation on the conditions for the determination as to whether or not the growth process is to be performed. In addition, the growth treatment and the apoptosis treatment may be performed as a set, or it may be individually determined whether or not to perform.

(Step S607) The growth unit 41 performs soma growth processing. An example of soma growth processing will be described using the flowchart of FIG.

(Step S608) The growth unit 41 performs AXON growth processing. An example of AXON growth processing will be described with reference to the flowchart of FIG.

(Step S609) The growth unit 41 performs Dendrites growth processing. An example of Dendrites growth processing will be described using the flowchart of FIG.

(Step S610) The growth unit 41 performs soma combining processing. An example of the soma combining process will be described with reference to the flowchart of FIG.

(Step S611) The growth unit 41 performs glial cell growth processing. An example of glial cell growth processing will be described using the flowchart of FIG.

(Step S612) The apoptosis processing unit 42 performs the apoptosis processing. It returns to step S601. An example of the apoptosis treatment will be described using the flowchart of FIG.

In the flowchart of FIG. 6, it is preferable that the firing transmission process, the firing pattern process, the growth process, the apoptosis process, etc. be processed in parallel. In addition, it is preferable that the firing transfer process of each soma is also processed in parallel.

Further, in the flowchart of FIG. 6, the processing is ended by the interruption of the power off or the processing end.

Next, details of the example of the ignition transmission process of step S603 will be described using the flowchart of FIG.

(Step S701) The feature information acquisition unit 611 acquires one or more feature information. The one or more pieces of feature information here are, for example, feature amounts of sound information. The method for the feature information acquisition unit 611 to acquire one or more pieces of feature information is, for example, any of the following methods (1) to (3). (1) The feature information acquisition unit 611 analyzes the input information received in step S602, and acquires one or more feature information. Here, the one or more pieces of feature information are, for example, one or more pairs of an information identifier and an information amount. (2) The information transfer means 612 acquires one or more pieces of feature information applied to the soma related information of the soma determined to be ignited. Here, the one or more pieces of feature information are, for example, one information amount. (3) The information transfer means 612 acquires one or more pieces of feature information from one or more pieces of feature information applied to the soma related information of the soma determined to be ignited. In (3), the information transfer means 612 calculates one or more pieces of feature information received by the soma determined to be fired according to a predetermined arithmetic expression to obtain one or more pieces of feature information. An example of such an operation is, for example, addition.

(Step S702) The information transfer means 612 determines one or more soma to which the one or more feature information acquired in step S701 is to be delivered. The determination method of such soma is, for example, any of the following (1) to (3). (1) The information transfer means 612 acquires one or more predetermined soma soma identifiers of the soma related information of the soma related information storage unit 51. The predetermined one or more soma soma identifiers are stored, for example, in the storage unit 5, and the information transfer unit 612 acquires the one or more soma identifiers from the storage unit 5. The one or more soma identifiers are soma's soma identifiers that first receive one or more pieces of feature information acquired from input information received from the outside. (2) The information transfer unit 612 acquires, from the firing start point information storage unit 55, one or more soma identifiers paired with the information identifier of the feature information acquired by the feature information acquisition unit 611. (3) The information transfer unit 612 refers to the combined information storage unit 53, and acquires one or more soma identifiers of one or more soma that are coupled to the end of the soma to be processed. That is, the information transfer means 612 realizes the processing of (3), for example, by the processing of any of the following (a) to (c). (A) The information transfer means 612 acquires, for example, from the combined information storage unit 53, Dendrites identifiers that are paired with one or more AXON identifiers included in soma related information of the soma to be processed. Then, the information transfer means 612 acquires, from the soma related information storage unit 51, one or more soma identifiers included in the soma related information having the acquired one or more Dendrites identifiers, for example. (B) The information transfer means 612 acquires, for example, one or more soma identifiers paired with the soma identifier included in the soma related information of the soma to be processed from the combined information storage unit 53. (C) The information transfer means 612 acquires, for example, from the combined information storage unit 53, a spine identifier paired with one or more synapse identifiers included in the soma related information of the soma to be processed. Then, the information transfer unit 612 acquires, for example, one or more soma identifiers included in the soma related information having the acquired one or more spine identifiers from the soma related information storage unit 51.

(Step S703) The processing unit 6 substitutes 1 into the counter i.

(Step S704) The processing unit 6 determines whether or not the i-th soma exists in the one or more soma determined in step S702. If the i-th soma exists, the process goes to step S 705. If the i-th soma does not exist, the process returns to the upper-level processing. The processing unit 6 normally determines whether or not the i-th soma identifier of the soma identifiers obtained in step S702 exists, as to whether or not the i-th soma exists.

(Step S 705) The determination unit 614 performs a firing determination process. The details of the firing determination process will be described with reference to the flowchart of FIG.

(Step S706) The processing unit 6 goes to step S707 if the judgment result in step S705 is “jump”, and goes to step S711 if the judgment result is “not fire”.

(Step S 707) The firing information storage unit 66 constructs firing information having a soma identifier identifying the soma determined by the determination means 614 to fire. Then, the firing information storage unit 66 stores the firing information in the firing information storage unit 59. Here, it is preferable that the firing information storage unit 66 acquire timer information from a timer (not shown), configure firing information having a soma identifier and timer information, and store the information in the firing information storage unit 59.

(Step S 708) The firing probability changing unit 615 changes the firing probability information paired with the soma identifier of the soma determined by the determining unit 614 to fire so that the probability of firing increases. The firing probability information to be changed is the information stored in the soma related information storage unit 51.

(Step S709) The processing unit 6 determines whether or not the transmission of the feature information to the soma ahead of the i-th soma or the soma group beyond the i-th soma is ended. When the transmission is finished, the process goes to step S711, and when the transmission is not finished, the process goes to step S710. The processing unit 6 refers to, for example, the combined information storage unit 53, and determines whether there is a soma ahead of the ith soma. When there is no previous soma of the i-th soma, the transmission of feature information is ended. In addition, if there is no AXON's destination Dendrites connected to the i-th soma, an action to transmit information from soma to AXON will be performed, but since there is no receiver of information from AXON onwards, Transmission stops. Also in this case, the processing unit 6 determines to end the transmission of the feature information.

(Step S710) The processing unit 6 executes an ignition transmission process to the soma ahead of the i-th soma or the soma group beyond the i-th soma. Such processing is ignition transmission processing. That is, the ignition transmission process is a recursive process.

(Step S711) The processing unit 6 increments the counter i by one. It returns to step S704.

Next, the details of the firing determination process of step S705 will be described using the flowchart of FIG.

(Step S801) The determination unit 614 acquires, from the soma related information storage unit 51, the firing condition information included in the soma related information of the soma to be processed.

(Step S802) The determination unit 614 determines whether external information is received in step S301. If the external information is received, the process goes to step S 803. If the external information is not received, the process goes to step S 804.

(Step S803) The determination unit 614 acquires the external information accepted in step S302 or one or more pieces of feature information acquired from the external information. Note that the process of acquiring one or more pieces of feature information from the external information may be performed by the feature information acquisition unit 611 or may be performed by the determination unit 614 or the like.

(Step S804) The determination unit 614 applies one or more pieces of feature information and the like to the firing condition information acquired in step S801, and determines whether or not to fire. The one or more feature information and the like are one or more feature information, or one or more feature information and external information.

(Step S805) The determination unit 614 substitutes the determination result in step S804 into a variable “return value”. Return to upper level processing.

Needless to say, it is not necessary to use external information in the flowchart of FIG.

Next, the details of the example of the firing pattern process of step S605 will be described using the flowchart of FIG.

(Step S901) The firing pattern acquisition unit 616 refers to the firing information storage unit 59 and obtains a firing pattern having one or more soma identifiers. In addition, it is preferable that the firing pattern acquisition unit 616 also acquires, from the learning information storage unit 58, a firing pattern corresponding to the input information or one or more pieces of feature information acquired from the input information. That is, it is preferable to use the firing pattern learned in association with the input information etc. in the following processing.

(Step S902) The output information acquisition unit 617 determines whether external information is received in step S301. If the external information is received, the process goes to step S 903. If the external information is not received, the process goes to step S 904.

(Step S903) The output information acquisition unit 617 acquires the external information accepted in step S301 or one or more feature information acquired from the external information. The process of acquiring one or more pieces of feature information from the external information may be performed by the feature information acquisition unit 611 or may be performed by the output information acquisition unit 617 or the like. The external information accepted in step S301 is usually external information stored in the storage unit 5.

(Step S904) The output information acquisition unit 617 substitutes 1 into the counter i.

(Step S 905) The output information acquisition unit 617 determines whether the i-th output management information exists in the output management information storage unit 56. If the i-th output management information exists, the process goes to step S906, and if the i-th output management information does not exist, the process goes to step S911.

(Step S 906) The output information acquisition unit 617 acquires the output condition of the i-th output management information from the output management information storage unit 56.

(Step S 907) The output information acquisition unit 617 determines whether the firing pattern acquired in step S 901 or the ignition pattern acquired in step S 901 and the information acquired in step S 903 match the output condition acquired in step S 906. To judge. If the output condition is met, the process goes to step S 908. If the output condition is not met, the process goes to step S 910.

(Step S908) The output information acquisition unit 617 acquires the output information included in the i-th output management information.

(Step S909) The output unit 4 outputs the output information acquired in step S908. The output here may be temporary accumulation of output information in a buffer. In such a case, the output unit 4 may output output sound information composed of two or more pieces of output information. The output sound information is, for example, output sound information obtained by combining (adding) two or more pieces of output information.

(Step S 910) The output information acquisition unit 617 increments the counter i by one. It returns to step S905.

(Step S 911) The learning information storage unit 63 acquires input information or one or more pieces of feature information acquired from the input information. Here, the input information is usually sound information.

(Step S912) The learning detection unit 62 substitutes 1 into the counter i.

(Step S 913) The learning detection unit 62 determines whether the i-th learning condition exists in the learning condition storage unit 57. If the i-th learning condition exists, the process goes to step S914. If the i-th learning condition does not exist, the process goes to step S917.

(Step S914) The learning detection unit 62 determines whether the firing pattern acquired in step S901 matches the i-th learning condition. If they match, the process goes to step S915. If they do not match, the process goes to step S917.

(Step S915) The learning information storage unit 63 acquires one or more soma identifiers to be stored from the firing pattern acquired in step S901. Then, the learning information storage unit 63 configures learning information including the one or more soma identifiers and the input information or the one or more feature information acquired in step S911. The one or more soma identifiers are firing patterns.

(Step S916) The learning information storage unit 63 stores the learning information configured in step S915.

(Step S917) The learning detection unit 62 increments the counter i by one. It returns to step S913.

In the flowchart of FIG. 9, steps S901 to S910 are output processing of output information, and steps S911 to S917 are learning processing.

Further, it goes without saying that external information may not be used in the flowchart of FIG.

Next, an example of the soma growth process of step S 607 will be described using the flowchart of FIG.

(Step S1001) The growth unit 64 substitutes 1 into the counter i.

(Step S 1002) The growth unit 64 determines whether the ith soma related information exists in the soma related information storage unit 51. If the i-th soma related information exists, the process goes to step S1003, and if the i-th soma related information does not exist, the process returns to the upper level processing.

(Step S1003) The growth unit 64 acquires, from the firing information storage unit 59, firing information corresponding to the i-th soma identifier included in the i-th soma related information. Such firing information is a history of firing of the ith soma. Then, the growth unit 64 determines whether the acquired firing information satisfies the condition. If the condition is satisfied, the process goes to step S1004, and if the condition is not satisfied, the process goes to step S1008. Note that the growing unit 64 determines that the condition is satisfied, for example, if soma is firing more appropriately as the condition determined in advance is satisfied from the acquired firing information. That is, the growing unit 64 determines whether the number of pieces of firing information including the ith soma identifier is large enough to satisfy a predetermined condition. If the number is large, it is determined that the condition is satisfied. The firing information acquired by the growth unit 64 may be, for example, firing information having timer information within a certain period from the present.

(Step S1004) The growth unit 64 acquires information on glial cell information corresponding to the ith soma. Here, information on glial cell information may be glial cell information of one or more glial cells coupled to the ith soma, or may be the number of glial cells coupled to the ith soma.

(Step S1005) The growth unit 64 determines whether the information on glial cell information acquired in step S1004 satisfies a predetermined condition. If the condition is satisfied, the process goes to step S1006. If the condition is not satisfied, the process goes to step S1008. The predetermined condition is a condition for dividing soma. Conditions for splitting soma are conditions for generating split soma. In addition, the predetermined condition is a case where the information on glial cell information is information indicating that the number of glial cells is so large that the predetermined condition is satisfied. The growth unit 64 calculates, for example, the number of glial cell identifiers paired with the i-th soma identifier, and determines that the predetermined condition is satisfied when the number is equal to or greater than the threshold or greater than the threshold. Here, the growth unit 64 is a condition on the condition that the number of one or more elements among the number of soma related information, the number of AXON information, the number of Dentrites information, etc. is smaller than the threshold or smaller than the threshold. It may be determined that

(Step S1006) The growth unit 64 generates soma related information of the split soma obtained by dividing the i-th soma, and stores the information in the soma related information storage unit 51. Note that there is no limitation on the algorithm by which the growing part 64 generates soma related information of the split soma.

(Step S1007) The growth unit 64 generates connection information for connecting the i-th soma and the split soma generated in step S1006, and stores the generated connection information in the connection information storage unit 53.

(Step S1008) The growth unit 64 increments the counter i by one. It returns to step S1002.

Next, an example of AXON growth processing in step S608 will be described using the flowchart of FIG.

(Step S1101) The growth unit 64 substitutes 1 into the counter i.

(Step S1102) The growth unit 64 determines whether or not the ith AXON information exists in the soma related information storage unit 51. If the i-th AXON information exists, the process goes to step S1103. If the i-th AXON information does not exist, the process returns to the upper level processing.

(Step S1103) The growth unit 64 acquires information on glial cell information corresponding to the i-th AXON information. Here, the information on glial cell information may be glial cell information of one or more glial cells coupled to the i-th AXON, or may be the number of glial cells coupled to the i-th AXON.

(Step S1104) The growth unit 64 determines whether the information on glial cell information acquired in step S1103 satisfies a predetermined condition. If the condition is satisfied, the process goes to step S1105. If the condition is not satisfied, the process goes to step S1106. The predetermined condition is a condition for AXON to be expanded. In addition, the predetermined condition is a case where the information on glial cell information is information indicating that the number of glial cells is so large that the predetermined condition is satisfied.

(Step S1105) The growth unit 64 changes AXON position information included in the i-th AXON information so that AXON can be expanded.

(Step S1106) The growth unit 64 increments the counter i by one. It returns to step S1102.

Next, Dendrites growth processing in step S609 will be described using the flowchart in FIG.

(Step S1201) The growth unit 64 substitutes 1 into the counter i.

(Step S1202) The growth unit 64 determines whether the ith Dendrites information exists in the soma related information storage unit 51 or not. If the i-th Dendrites information exists, the process goes to step S1203, and if the i-th Dendrites information does not exist, the process returns to the upper level processing.

(Step S1203) The growth unit 64 acquires information on glial cell information corresponding to the ith Dendrites information. Here, information on glial cell information may be glial cell information of one or more glial cells coupled to the ith Dendrites, or may be the number of glial cells coupled to the ith Dendrites.

(Step S1204) The growth unit 64 determines whether the information on the glial cell information acquired in step S1203 satisfies a predetermined condition. If the condition is satisfied, the process proceeds to step S1205. If the condition is not satisfied, the process proceeds to step S1206. The predetermined condition is a condition for extending Dendrites. In addition, the predetermined condition is a case where the information on glial cell information is information indicating that the number of glial cells is so large that the predetermined condition is satisfied.

(Step S1205) The growing unit 64 changes Dendrites position information included in the i-th Dendrites information so that Dentrites is expanded.

(Step S1206) The growth unit 64 increments the counter i by one. It returns to step S1202.

Next, the soma combining process of step S610 will be described using the flowchart of FIG.

(Step S1301) The growth unit 64 substitutes 1 into the counter i.

(Step S1302) The growth unit 64 determines whether or not the ith AXON information exists in the soma related information storage unit 51. If the i-th AXON information exists, the process goes to step S1303, and if the i-th AXON information does not exist, the process returns to the upper level processing.

(Step S1303) The growth unit 64 acquires AXON position information included in the i-th AXON information from the soma related information storage unit 51.

(Step S1304) The growth unit 64 substitutes 1 into the counter j.

(Step S1305) The growing unit 64 is Dendrites information of Dendrites which is not an input to the soma to which the i-th AXON is connected, and whether the j-th Dendrites information exists in the soma related information storage unit 51 or not to decide. If the j-th Dendrites information exists, the process goes to step S1306, and if the j-th Dendrites information does not exist, the process goes to step S1310.

(Step S1306) The growth unit 64 acquires Dendrites position information included in the j-th Dendrites information from the soma related information storage unit 51.

(Step S1307) Whether the growing section 64 can combine the i-th AXON with the j-th Dendrites using the AXON position information included in the i-th AXON information and the Dendrites position information included in the j-th Dendrites information Decide whether or not. If the combination is possible, the process goes to step S1308. If the combination is not possible, the process goes to step S1309. When it is determined that, for example, the AXON position information included in the i-th AXON information overlaps the Dendrites position information included in the j-th Dendrites information, the growing unit 64 determines that the i-th AXON and the j-th Dendrites Determines that it is possible to combine. In the growing unit 64, for example, the position of the tip of AXON indicated by AXON position information included in the i-th AXON information matches the position of the tip of Dendrites indicated by Dendrites position information included in the j-th Dendrites information. Alternatively, when it is determined that the distance is equal to or less than the threshold value, it is determined that the i-th AXON and the j-th Dendrites can be combined.

(Step S1308) The growth unit 64 constructs connection information for specifying the connection between the i-th AXON and the j-th Dendrites, and stores the connection information in the connection information storage unit 53. The growing unit 64 constructs, for example, combination information including an AXON identifier included in the i-th AXON information and a Dendrites identifier included in the j-th Dendrites information, and stores the combined information in the combined information storage unit 53.

(Step S1309) The growth unit 64 increments the counter j by one. It returns to step S1305.

(Step S1310) The growth unit 64 increments the counter i by one. It returns to step S1302.

Next, the glial cell growth process of step S611 will be described using the flowchart of FIG.

(Step S1401) The growth unit 64 substitutes 1 into the counter i.

(Step S1402) The growth unit 64 determines whether the ith soma related information exists in the soma related information storage unit 51. If the i-th soma related information exists, the process goes to step S1403, and if the i-th soma related information does not exist, the process returns to the upper process.

(Step S1403) The growth unit 64 acquires the necessary energy amount information and the possessed energy amount information that the i-th soma related information has.

(Step S1404) The growth unit 64 determines whether the necessary energy amount information and the possessed energy amount information acquired in step S1403 satisfy the predetermined condition. If the predetermined condition is satisfied, the process proceeds to step S1405. If the predetermined condition is not satisfied, the process proceeds to step S1407.

(Step S1405) The growth part 64 acquires the soma identifier which the i-th soma related information has.

(Step S1406) The growth unit 64 constructs glial cell information having the soma identifier acquired in step S1405, and stores the glial cell information in the glial cell information storage unit 54.

(Step S1407) The growth unit 64 substitutes 1 into the counter j.

(Step S1408) The growth unit 64 determines whether the j-th AXON information exists in the i-th soma related information. If the j-th AXON information exists, the process goes to step S1409, and if the j-th AXON information does not exist, the process goes to step S1414.

(Step S1409) The growth unit 64 acquires necessary energy amount information and possessed energy amount information included in the j-th AXON information.

(Step S1410) The growth unit 64 determines whether the necessary energy amount information and the possessed energy amount information acquired in step S1409 satisfy the predetermined conditions. If the predetermined condition is satisfied, the process goes to step S1411, and if the predetermined condition is not satisfied, the process goes to step S1413.

(Step S1411) The growth unit 64 acquires the AXON identifier of the j-th AXON information.

(Step S1412) The growth unit 64 constructs glial cell information having the AXON identifier acquired in step S1411, and stores the glial cell information in the glial cell information storage unit 54.

(Step S1413) The growth unit 64 increments the counter j by one. It returns to step S1408.

(Step S1414) The growth unit 64 substitutes 1 into the counter j.

(Step S1415) The growth unit 64 determines whether the j-th Dendrites information is present in the i-th soma related information. If the j-th Dendrites information exists, the process goes to step S1409, and if the j-th Dendrites information does not exist, the process goes to step S1414.

(Step S1416) The growth unit 64 acquires necessary energy amount information and possessed energy amount information included in the j-th Dendrites information.

(Step S1417) The growth unit 64 determines whether the necessary energy amount information and the possessed energy amount information acquired in step S1416 satisfy the predetermined conditions. If the predetermined condition is satisfied, the process goes to step S1418. If the predetermined condition is not satisfied, the process goes to step S1420.

(Step S1418) The growth unit 64 acquires a Dendrites identifier of the j-th Dendrites information.

(Step S1419) The growth unit 64 constructs glial cell information having the Dendrites identifier acquired in step S1418, and stores the glial cell information in the glial cell information storage unit 54.

(Step S1420) The growth unit 64 increments the counter j by one. It returns to step S1415.

(Step S1421) The growth unit 64 increments the counter i by one. It returns to step S1402.

Next, the apoptosis process of step S612 will be described using the flowchart of FIG.

(Step S1501) The apoptosis processing unit 65 substitutes 1 into the counter i.

(Step S1502) The apoptosis processing unit 65 determines whether the ith soma related information exists in the soma related information storage unit 51. If the i-th soma related information exists, the process goes to step S1503. If the i-th soma related information does not exist, the process returns to the upper processing.

(Step S1503) The apoptosis processing unit 65 acquires the soma identifier of the ith soma related information from the soma related information storage unit 51.

(Step S1504) The apoptosis processing unit 65 acquires, from the ignition information storage unit 59, the ignition information including the soma identifier acquired in step S1503. Preferably, the firing information acquired here is firing information having timer information indicating a time within the threshold or closer to the threshold than the current time.

(Step S1505) The apoptosis processing unit 65 determines, using the firing information acquired in step S1504, whether the condition for apoptosis is satisfied. If the condition for apoptosis is satisfied, the process goes to step S1506. If the condition for apoptosis is not satisfied, the process goes to step S1508.

(Step S1506) The apoptosis processing unit 65 deletes the ith soma related information from the soma related information storage unit 51.

(Step S1507) The apoptosis processing unit 65 deletes the binding information corresponding to the ith soma from the binding information storage unit 53.

(Step S1508) The apoptosis processing unit 65 increments the counter i by one. It returns to step S1502.

The software for realizing the sound information processing apparatus in the present embodiment is the following program. That is, this program divides the computer into a reception unit that receives sound information sequentially, and divides the sound information received by the reception unit into two or more frequency bands, and acquires two or more partial sound information The program is configured to function as a changing unit that synthesizes two or more partial sound information having different frequency bands at different timings to obtain output sound information, and an output unit that outputs the output sound information.

In the above program, a computer-accessible recording medium is a soma related information in which two or more soma related information having a soma identifier for identifying soma and firing condition information on a condition for firing the soma is stored A condition for output using an information storage unit, a combined information storage unit storing one or more combined information specifying a combination between two or more soma, and a firing pattern having one or more soma identifier And an output management information storage unit storing one or more output management information having output conditions and output information which is output information, and the change unit is configured to receive two or more pieces of sound information received by the reception unit. First division means for acquiring two or more pieces of first partial sound information, and temporally dividing the sound information received by the reception unit into two or more; Second division means for acquiring two or more pieces of second partial sound information, and characteristic information acquisition means for acquiring one or more pieces of characteristic information from partial sound information that is the first partial sound information or the second partial sound information; Acquiring one or more soma identifiers identifying the soma to be fired first for each partial sound information, and one or more feature information passed from one or more other soma or the one or more feature information An information transfer means for obtaining one or more feature information to be obtained and a soma identifier of one or more soma to be subjected to the judgment of firing, and the information transfer means for each of the two or more pieces of feature information It is determined whether or not the soma identified by each of the soma identifiers is fired using the one or more feature information and the firing condition information paired with the one or more soma identifiers acquired by the information transfer means A firing pattern acquiring unit for acquiring a firing pattern including one or more soma identifiers for identifying the soma determined to fire by the partial sound information, and for each of the partial sound information, Computer comprising: output means corresponding to a firing pattern obtained by the firing pattern obtaining means, obtained from the output management information storage unit, and synthesizing means for composing output sound information from the obtained one or more of the output information It is preferable that it is a program that causes

FIG. 16 also shows the appearance of a computer that implements the sound information processing apparatus of the various embodiments described above by executing the program described in the present specification. The embodiments described above can be implemented with computer hardware and computer programs executed thereon. FIG. 16 is an overview of this computer system 300, and FIG. 17 is a block diagram of the system 300.

In FIG. 16, computer system 300 includes a computer 301 including a CD-ROM drive, a keyboard 302, a mouse 303, and a monitor 304.

In FIG. 17, a computer 301 includes an MPU 3013, a bus 3014 connected to the MPU 3013, and a CD-ROM drive 3012 in addition to a CD-ROM drive 3012, a ROM 3015 for storing programs such as a bootup program, and an MPU 3013. And a RAM 3016 for temporarily storing instructions of the application program and providing a temporary storage space, and a hard disk 3017 for storing the application program, the system program, and data. Although not shown here, the computer 301 may further include a network card for providing a connection to the LAN.

The program that causes the computer system 300 to execute the functions of the sound information processing apparatus of the above-described embodiment may be stored in the CD-ROM 3101, inserted into the CD-ROM drive 3012, and further transferred to the hard disk 3017. Alternatively, the program may be transmitted to the computer 301 via a network (not shown) and stored in the hard disk 3017. The program is loaded into the RAM 3016 upon execution. The program may be loaded directly from the CD-ROM 3101 or from the network.

The program may not necessarily include the operating system (OS) or the third party program that causes the computer 301 to execute the function of the sound information processing apparatus of the above-described embodiment. The program needs to call only an appropriate function (module) in a controlled manner, and include only a part of an instruction that makes it possible to obtain a desired result. It is well known how the computer system 300 operates, and the detailed description is omitted.

In the above program, in the step of transmitting information, the step of receiving information, etc., processing performed by hardware, for example, processing performed by modem or interface card in transmission step (performed only by hardware Processing is not included.

Moreover, the computer that executes the program may be singular or plural. That is, centralized processing may be performed, or distributed processing may be performed.

Further, in each of the above embodiments, it goes without saying that two or more communication means existing in one apparatus may be physically realized by one medium.

In each of the above embodiments, each process may be realized by centralized processing by a single device, or may be realized by distributed processing by a plurality of devices.

It goes without saying that the present invention is not limited to the above embodiments, and various modifications are possible, which are also included in the scope of the present invention.

As described above, the sound information processing apparatus according to the present invention has the effect of being able to obtain sound information obtained by imitating the function of the human ear or brain, and is useful as a sound information processing apparatus etc. is there.

1, 5 storage unit 2 reception unit 3, 61 change unit 4 output unit 6 processing unit 21 input information reception unit 31 first division unit 32

second division unit

33, 617 synthesis unit 51 soma related information storage unit 52 soma group information storage Unit 53 Connection information storage unit 54 Glial cell information storage unit 55 Firing start point information storage unit 56 Output management information storage unit 57 Learning condition storage unit 58 Learning information storage unit 59 Ignition information storage unit 60 Usage connection information storage unit 62 Learning detection unit 63 Learning information storage unit 64 Growth unit 65 Apoptosis processing unit 66 Firing information storage unit 611 Characteristic information acquisition means 612 Information transmission means 613 soma arithmetic means 614 Judgment means 615 Firing probability change means 616 Firing pattern acquisition means 617 Output information acquisition means 618 Control means 6121 Firing start point soma determination means 6122 Connection detection Means 6123 Transmission information acquisition means

Claims

A reception unit that receives sound information sequentially;
The sound information received by the reception unit is divided into two or more frequency bands, two or more partial sound information is acquired, and two or more partial sound information with different frequency bands at different timings are synthesized and output. A change unit that acquires sound information,
A sound information processing apparatus comprising: an output unit that outputs the output sound information.
The change unit is
The sound information received by the reception unit is divided into two or more frequency bands, two or more partial sound information is acquired, and partial sound information in a high frequency band of the sound information received earlier in time, and time The sound information processing apparatus according to claim 1, wherein the sound information processing apparatus acquires the output sound information by combining the partial sound information in the low frequency band among the sound information received later.
The change unit is
First division means for dividing the sound information received by the reception unit into two or more frequency bands, acquiring two or more first partial sound information, and temporarily storing the first partial sound information in a first queue;
Second division means for temporally dividing the sound information received by the reception unit into two or more, acquiring two or more pieces of second partial sound information, and temporarily storing the second partial sound information in a second queue;
A combining means for sequentially extracting and combining the one or more first partial sound information of the first cue and the one or more second partial sound information of the second cue; and acquiring output sound information The sound information processing apparatus according to claim 1 or 2.
a soma related information storage unit storing two or more pieces of soma related information including a soma identifier identifying soma and firing condition information on a condition for firing the soma;
A combined information storage unit in which one or more pieces of combined information specifying a combination between two or more soma are stored;
An output management information storage unit storing one or more pieces of output management information having output conditions which are conditions for output using a firing pattern having one or more soma identifiers and output information which is output information; In addition,
The change unit is
First division means for dividing the sound information received by the reception unit into two or more frequency bands and acquiring two or more first partial sound information;
Second division means for temporally dividing the sound information received by the reception unit into two or more, and acquiring two or more pieces of second partial sound information;
Feature information acquisition means for acquiring one or more pieces of feature information from the partial sound information which is the first partial sound information or the second partial sound information;
For each partial sound information, obtain one or more soma identifiers that identify the soma to be fired first, and obtain from one or more feature information passed from one or more other soma or the one or more feature information Information transmission means for acquiring one or more feature information to be detected and one or more soma identifiers of each soma to be subjected to the judgment of firing;
For each of the two or more pieces of feature information, using the one or more pieces of feature information acquired by the information transfer means, and the firing condition information paired with the one or more soma identifiers obtained by the information transfer means Determination means for determining whether or not the soma identified by each soma identifier fires;
An ignition pattern acquisition unit that acquires, for each of the partial sound information, an ignition pattern that includes one or more soma identifiers that identify the soma that the determination unit has determined to fire;
A synthesis unit that acquires output information corresponding to the firing pattern acquired by the ignition pattern acquisition unit from the output management information storage unit for each of the partial sound information, and configures output sound information from the acquired one or more output information The sound information processing apparatus according to any one of claims 1 to 3, comprising:
A sound information processing method implemented by a receiving unit, a changing unit, and an output unit,
An accepting step in which the accepting unit sequentially accepts sound information;
The changing unit divides the sound information received in the receiving step into two or more frequency bands, acquires two or more partial sound information, and two or more partial sounds having different frequency bands at different timings. Changing step of synthesizing information and acquiring output sound information;
A sound information processing method comprising: an output step of outputting the output sound information by the output unit.
Computer,
A reception unit that receives sound information sequentially;
The sound information received by the reception unit is divided into two or more frequency bands, two or more partial sound information is acquired, and two or more partial sound information with different frequency bands at different timings are synthesized and output. A change unit that acquires sound information,
A program for functioning as an output unit that outputs the output sound information.