JP2012223314A - Loudness measuring apparatus, hearing-aid adjusting device, method and program thereof - Google Patents

Abstract

PROBLEM TO BE SOLVED: To solve the problem that it takes time for measuring a loudness increase curve.SOLUTION: A method for measuring loudness presents a plurality of initial presentation sounds having a different frequencies with each other and sound pressure with a same loudness value, measures brain waves of a user when the initial presentation sound is presented, selects a similar typical pattern while expanding the typical pattern of brain waves including an event-related potential stored in a typical pattern storage part to compare the typical pattern with the event-related potential of the measured brain waves, presents a plurality of presentation sounds having a different frequencies with each other and different sound pressure, measures the brain waves of the user when the presentation sounds are presented, and acquires an N1 component included in the typical pattern of the brain waves expanded so as to be most similar to the event-related potential relative to the plurality of presentation sounds while expanding the selected typical pattern, and obtains the loudness increase curve referring to the relation between latency of a predetermined N1 component and loudness and using the latency of the obtained N1 component.

Description

  The present invention relates to an apparatus and a method for measuring a sensory sound volume (loudness) received by a human relative to a physical sound volume.

  There is a case where there is not a one-to-one correspondence between the sound pressure level of sound and the level of sensory sound received by humans. In this specification, “loudness” indicates the loudness of a sensory sound that a human receives.

  The loudness of the aurally-challenged is different from that of a normal hearing person. For example, many hearing-impaired persons cannot hear a sound with a low sound pressure that can be heard by a normal hearing person. However, many hearing-impaired persons can hear a sound with a high sound pressure, like a normal hearing person. The loudness of a hearing impaired person is distorted compared to the loudness of a normal hearing person. This phenomenon is referred to as a recruitment phenomenon.

  Consider a case where a hearing aid used by a hearing-impaired person is adjusted to increase the sound pressure uniformly for all sound pressure ranges. This adjustment allows the hearing impaired to hear a sound with a low sound pressure. On the other hand, due to the adjustment, a hearing impaired person feels that a sound with a large sound pressure is too loud.

  Therefore, in order to adjust the hearing aid, it is necessary to measure the loudness including the supplementary phenomenon. It is required to adjust the magnitude of sound pressure using the measured loudness.

  There is a conventional technique for measuring human hearing using brain waves induced when a human hears sound. The prior art shown in Non-Patent Document 1 uses an auditory brain-stem response (ABR) to determine whether an infant has a hearing impairment. Specifically, when the infant hears a sound having a specific sound pressure, it is determined whether the infant's brain wave waveform is a typical ABR waveform. The determination result corresponds to the determination of whether the infant has hearing impairment.

  The prior art shown in Patent Document 1 performs an hearing test using an auditory evoked potential (AEP). Specifically, for example, a reference stimulus having 2 kHz and 90 dBHL is presented, and the AEP is measured. Hearing is inspected based on a threshold value using the measured AEP.

Special table 2010-504139

P.R.Kileny “New insight on infant ABR hearing screening” Scandinavian Audiology Supplement 30, 81-88, 1988

  In order to measure the loudness, it is necessary to acquire an N1 component included in an electroencephalogram described later. When the N1 component is acquired only from the measured electroencephalogram, it is necessary to measure a very large amount of electroencephalogram signals, which is a burden on the user.

  In order to reduce the number of electroencephalograms to be measured, there is a method of comparing a typical ABR waveform and a measured electroencephalogram, as in the prior art shown in Non-Patent Document 1. In addition, there is a method of measuring an electroencephalogram serving as a reference based on a reference stimulus as in the prior art disclosed in Patent Document 1.

  However, since the N1 component is different for each sound frequency, sound pressure, and individual, it is not easy to obtain the N1 component using a uniform standard.

  Therefore, the present invention provides an apparatus and method for measuring loudness including a supplementary phenomenon of a user with a smaller amount of brain waves than in the prior art by using knowledge about changes in brain waves with respect to sound frequency and sound pressure. For the purpose.

  In order to solve the above-described conventional problems, the present invention provides a plurality of initial presentation sounds having sound pressures that are different from each other and have the same loudness that is the loudness of a sensory sound received by humans. The step of presenting to the user, the step of measuring the brain wave of the user when the initial presentation sound is presented, and the expansion and contraction of the typical pattern of the electroencephalogram including the event-related potential stored in the typical pattern storage unit, A typical pattern selection step of comparing a typical pattern and an event-related potential after a predetermined time from the time of the plurality of initial presentation sounds among the measured electroencephalograms, and selecting a typical pattern similar to the event-related potential; and A step of presenting to the user a plurality of present sounds having different frequencies and different sound pressures; and a step of measuring the user's brain wave when the present sound is presented. And the N1 component included in the typical pattern expanded and contracted so as to be most similar to the event-related potential after a predetermined time from the time of the plurality of presenting sounds in the brain wave. When the electroencephalogram including the N1 component is measured with reference to the relationship between the EEG selection step to be performed and the relationship between the latency and the loudness of the N1 component determined in advance, and the loudness value corresponding to the acquired latency of the N1 component Obtaining a loudness increase curve as the sound pressure of the present sound.

  According to the apparatus and method of the present invention, an apparatus for measuring loudness including a supplementary phenomenon of a user with a smaller amount of brain waves than in the prior art by using knowledge about changes in brain waves with respect to sound frequency and sound pressure, and A method can be provided.

1 is a diagram illustrating a configuration of a loudness measurement system according to a first embodiment. The figure which shows the relationship between the sound pressure of a presentation sound, N1 latency, the amplitude of N1 component, and the amplitude difference of P2 component. Schematic diagram of the EEG waveform measured when the same sound was presented to multiple users FIG. 3 is a diagram illustrating a configuration of a portion of the loudness measurement system according to the first embodiment. FIG. 3 is a diagram illustrating a configuration of a portion of the loudness measurement system according to the first embodiment. 3 is a flowchart showing the operation of the loudness measurement system according to the first embodiment. 3 is a flowchart showing the operation of the loudness measurement system according to the first embodiment. 3 is a flowchart showing the operation of the loudness measurement system according to the first embodiment. The figure which shows the procedure which calculates | requires a loudness increase curve. FIG. 3 is a diagram illustrating a matching procedure of the loudness measurement system according to the first embodiment. 3 is a flowchart showing the operation of the loudness measurement system according to the first embodiment. 3 is a flowchart showing the operation of the loudness measurement system according to the first embodiment. FIG. 6 is a diagram showing a configuration of a loudness measurement system according to a first modification of the first embodiment. The figure which shows the relationship between the sound pressure which gives the minimum audible value and equal loudness in the modification 1 of Embodiment 1. FIG. The schematic diagram which shows the procedure which determines the input-output characteristic of the hearing aid in the modification 1 of Embodiment 1. FIG. The figure which shows the structure of the part of the loudness measurement system of the modification 2 of Embodiment 1. FIG. 7 is a flowchart showing the operation of the loudness measurement system according to the second modification of the first embodiment. The figure which shows the function for determining the deformation rate of the typical pattern in the loudness measurement system of the modification 2 of Embodiment 1. FIG. 7 is a flowchart showing the operation of the loudness measurement system according to the second modification of the first embodiment. FIG. 3 is a diagram illustrating a configuration of a loudness measurement system according to a second embodiment. 6 is a flowchart showing the operation of the loudness measurement system according to the second embodiment. The figure which shows the typical pattern of the evoked potential of the loudness measurement system of Embodiment 2. FIG. FIG. 4 is a diagram illustrating a configuration of a portion of a loudness measurement system according to a second embodiment. 6 is a flowchart showing the operation of the loudness measurement system according to the second embodiment. 6 is a flowchart showing the operation of the loudness measurement system according to the second embodiment. The figure which shows the electrode position of an international 10-20 method (10-20 System). The figure which shows the example of an electroencephalogram sample pattern. The figure which shows an example of an equal loudness curve. The figure which shows the hardware constitutions of a loudness measuring device.

(Definition)
Definitions of terms used in this specification will be described.
“Event-related potential (ERP)” is a type of electroencephalogram (EEG), which is a transient potential change in the brain that is temporally related to an external or internal event. Say. An “evoked potential” is an example of an event-related potential. The “auditory stimulus” is a sound presented to the user. The “N1 component” is an evoked potential of a negative component after a range of 50 ms or more and 150 ms or less starting from an auditory stimulus presentation. The N1 component is included in the event-related potential. The “P2 component” is an evoked potential of a positive component after a range of 150 ms or more and 300 ms or less starting from an auditory stimulus presentation. The P2 component is included in the event-related potential. The “latency” is the time until the peak potential of the positive component or negative component appears from the time when the voice stimulus is presented. “Negative component” generally refers to a potential lower than 0 μV. A “positive component” generally refers to a potential greater than 0 μV. “UCL (uncomfortable loudness level)” is a sound pressure that is so high that the user feels uncomfortable. The “minimum audible threshold level (HTL)” is the sound pressure of the smallest sound that the user can hear. “Present a sound” means outputting a pure sound. A “pure tone” is a sound represented by a sine wave that repeats periodic vibration at a single frequency.

  According to Table 1 described on page 30 of "Event-related potential (ERP) manual-mainly on P300" (edited by Kimitaka Kaga et al., Shinohara Publishing Shinsha, 1995), in general, In the waveform, a difference (shift) of 30 ms to 50 ms occurs for each individual. Therefore, in this specification, “about Xms” or “near Xms” includes a width of 30 to 50 ms centered on Xms (for example, 100 ms ± 30 ms, 200 ms ± 50 ms).

(Characteristics of latency of N1 component, change of brain wave to sound frequency and sound pressure)
Prior to the detailed description of the present embodiment, the characteristics of the latency of the N1 component and the knowledge about the change of the electroencephalogram with respect to the sound frequency and the sound pressure will be described.

  When the loudness of the sensory sound received by humans is large, the latency of the N1 component is short. When the loudness is small, the latency of the N1 component is long. FIG. 2 shows the relationship between the loudness of the sensory sound received by humans, the latency of the N1 component, and the amplitude of the N1 component. The relationship shown in FIG. 2 is shown in “Chapter 23 Slow Response to the Head”, edited by Sotaro Funasaka, edited by Shinjiro Onishi, supervised by Atsuro Suzuki (1985), “Auditory Brainstem Reaction—its Basics and Clinical—” 381-392. The horizontal axis in FIG. 2 is dBSL. The horizontal axis of FIG. 2 displays the sound pressure in dB with the listening threshold of the user's stimulation sound set to 0 dB. The display on the left side of the vertical axis in FIG. 2 indicates the magnitude of the N1-P2 amplitude. The unit of the magnitude of the N1-P2 amplitude is microvolts (μV). The display on the right side of the vertical axis in FIG. 2 indicates the latency of N1. The unit of N1 latency is milliseconds (ms). FIG. 2 shows the measurement results when a 1000 Hz presentation sound was presented to a normal hearing person as an auditory stimulus. Generally, it is known that UCL (uncomfortable loudness level) is about 90 dB to 100 dB for a sound having a frequency of 1000 Hz. That is, UCL is larger than 90 dBSL at the right end on the horizontal axis of FIG.

  White circles shown in FIG. 2 indicate latency data. The broken line shown in FIG. 2 shows the model value of the latency. Black circles shown in FIG. 2 indicate amplitude data. The solid line shown in FIG. 2 indicates the amplitude model value. In FIG. 2, when the sound pressure of the presentation sound increases from 10 dBSL to 90 dBSL, the amplitude increases relatively linearly. When the sound pressure of the presentation sound increases from 10 dBSL to 30 dBSL, the latency is sharply shortened. When the sound pressure of the presenting sound increases from 30 dBSL to 70 dBSL, the latency is relatively linearly shortened. As shown in FIG. 2, when the sound pressure of the presentation sound is large, the latency of the N1 component is short, and when the sound pressure of the presentation sound is small, the latency of the N1 component is long.

  FIGS. 3A to 3C show examples of electroencephalograms measured after presenting sound. FIGS. 3A to 3C are schematic diagrams of waveforms measured when the same sound is presented to a plurality of users. 3 (a) to 3 (c), the vertical axis represents potential and the horizontal axis represents time. Each waveform has an N1 component and a P2 component. In the waveform shown in FIG. 3A, the amplitude of the N1 component and the amplitude of the P2 component are approximately the same. On the other hand, in the waveform shown in FIG. 3B, the amplitude of the P2 component is larger than the amplitude of the N1 component. As described above, the ratio of the amplitude of the N1 component and the amplitude of the P2 component or the shape of the waveform of the electroencephalogram is different for each individual. Also, the electroencephalogram differs depending on the sound frequency and the sound pressure.

  However, there is a finding that even if the sound frequency and the sound pressure of the sound are changed, the change in the waveform between the N1 component and the P2 component included in the brain wave of the same user is small. The waveform between the N1 component and the P2 component is, for example, a portion surrounded by a broken line in the waveforms of FIGS. Therefore, when measuring the sound frequency and the sound pressure by changing the sound, the inventors of the present application change the shape of the waveform between the N1 component and the P2 component so as to approximately maintain the waveform for each individual. It was found that a typical pattern of suitable electroencephalograms can be selected.

  Hereinafter, embodiments of the present invention will be described with reference to the drawings.

(Embodiment 1)
FIG. 1 is a configuration diagram of a loudness measurement system 10 according to the first embodiment.

  The loudness measurement system 10 includes a presentation sound determination unit 110, a presentation sound generation unit 120, an electroacoustic conversion unit 130, an electroencephalogram measurement unit 150, a sample pattern generation unit 160, a typical pattern storage unit 170, and a typical pattern selection. Unit 180, brain wave selection unit 190, latency estimation unit 200, measurement completion determination unit 210, and presentation sound control unit 220. Each configuration of the loudness measurement system 10 is connected by wire or wirelessly. In FIG. 1, a user 100 shown for convenience of explanation is not a component of the loudness measurement system 10.

(Presentation sound determination unit 110)
The presentation sound determination unit 110 determines a plurality of sounds having different frequencies from the sounds stored in the presentation sound storage unit 111. “Presentation sound” is a sound presented to the user 100. The presentation sound determination unit 110 may determine a plurality of sounds from the sounds stored in the presentation sound storage unit 111 with reference to a plurality of predetermined frequencies.

  The presentation sound determination unit 110 has a loudness correspondence relationship between the value of the sound pressure and the value of the sensuous sound (loudness) received by humans for each frequency. This loudness correspondence is a so-called equal loudness curve. The equal loudness curve has a corresponding relationship with the sound pressure value at which the loudness is the same for each frequency. FIG. 28 shows an example of an equal loudness curve. The frequency and sound pressure on the curve shown in FIG. 28 have the same loudness.

  The presented sound determination unit 110 refers to the loudness correspondence, and determines sound pressures having the same loudness value for a plurality of presented sounds having different frequencies. For example, an arbitrary loudness value is determined. Next, with reference to the loudness correspondence, the sound pressure corresponding to the determined loudness value is determined for each frequency of the presentation sound. For example, when using the equal loudness curve shown in FIG. 28, a pure tone having 250 Hz and 50 dB (A in FIG. 28), a pure tone having 1000 Hz and 40 dB (B in FIG. 28), and a pure tone having 4000 Hz and 35 dB (C in FIG. 28). ).

  The presentation sound determination unit 110 transmits information on a plurality of presentation sounds to the presentation sound generation unit 120. The “presentation sound information” includes information on the frequency and sound pressure of the presentation sound.

(Presentation sound storage unit 111)
The presented sound storage unit 111 stores at least a plurality of pure sounds associated with frequencies. The presentation sound storage unit 111 may store a pure tone associated with the type of phoneme, the level of sound pressure, and the like.

(Presentation sound generation unit 120)
The presentation sound generation unit 120 receives information about a plurality of presentation sounds including information on the frequency and sound pressure of the presentation sound from the presentation sound determination unit 110. The presentation sound generation unit 120 determines the output order of the plurality of presentation sounds and the intervals at which the plurality of presentation sounds are output, using the received information of the plurality of presentation sounds. The presenting sound generation unit 120 may previously hold the number of times of presenting the presenting sound and the interval of presenting the presenting sound. Or the presentation sound production | generation part 120 may receive the frequency | count of presenting a presentation sound, and the space | interval which presents a presentation sound.

  The present sound generation unit 120 refers to the present sound information and the number of times the present sound is presented, and determines the order so that present sounds having the same frequency do not continue. Hereinafter, the sound presented to the user including the presenting sound is also referred to as “stimulus”. The interval for presenting the presenting sound is also referred to as “a interval between stimulus” or “presentation interval”.

  The presenting sound generation unit 120 transmits to the electroacoustic transducer 130 sound information of a presenting sound including information on a plurality of presenting sounds, the order of presenting the presenting sound, and an interval for presenting the presenting sound.

(Electroacoustic transducer 130)
The electroacoustic transducer 130 receives audio information of the presentation sound from the presentation sound generation unit 120 and the sound control unit 220. The electroacoustic transducer 130 converts the sound output to the user 100 into sound data corresponding to the sound information of the sound to be displayed. The electroacoustic transducer 130 transmits the time when the presentation sound is output to the sample pattern generation unit 160. For example, the electroacoustic transducer 130 includes a device that outputs sound including a speaker and headphones. The electroacoustic transducer 130 is also referred to as a “presentation unit”.

(Electroencephalogram measurement unit 150)
A plurality of electrodes 140 including a reference electrode 141 and a measurement electrode 142 are disposed on the user 100. The electroencephalogram measurement unit 150 measures an electroencephalogram corresponding to the potential difference between the reference electrode 141 and the measurement electrode 142. The electroencephalogram measurement unit 150 transmits the measured electroencephalogram to the sample pattern generation unit 160 and the electroencephalogram selection unit 190.

(Sample pattern generation unit 160)
The sample pattern generation unit 160 receives the measured electroencephalogram from the electroencephalogram measurement unit 150. The sample pattern generation unit 160 receives the time when the presentation sound is presented from the electroacoustic transducer 130. The sample pattern generation unit 160 extracts, from the measured electroencephalogram, an electroencephalogram in a range after a predetermined time from the time when the presentation sound is presented. The brain waves in a range after a predetermined time from the time when the presentation sound is presented are, for example, the N1 component and the P2 component. An electroencephalogram sample pattern is generated by adding the extracted electroencephalograms to sounds having different frequencies. The sample generation unit 160 transmits the generated electroencephalogram sample pattern to the typical pattern selection unit 180.

(Typical pattern storage unit 170)
The typical pattern storage unit 170 stores a plurality of typical patterns of an electroencephalogram including an N1 component and a P2 component. You may hold | maintain the electroencephalogram waveform itself matched with the time from a presentation sound. Moreover, you may hold | maintain the information of the electric potential value matched with the time from a presentation sound.

(Typical pattern selection unit 180)
The typical pattern selection unit 180 receives the brain wave sample pattern generated by the sample pattern generation unit 160. The typical pattern selection unit 180 detects the N1 component and the P2 component from the received brain wave sample pattern. The typical pattern selection unit 180 acquires the latency time of the N1 component, the latency time of the P2 component, and the difference between the latency of the N1 component and the latency of the P2 component.

  The typical pattern selection unit 180 determines a matching expansion / contraction range from the difference between the latency of the N1 component and the latency of the P2 component, and the magnification previously held. The typical pattern selection unit 180 determines the matching time range using the latency time of the N1 component, the latency time of the P2 component, and information on the time width stored in advance.

  The typical pattern selection unit 180 selects a typical pattern similar to the sample pattern of the electroencephalogram while expanding / contracting using the determined expansion / contraction range and time range. The typical pattern selection unit 180 transmits the selected typical pattern to the electroencephalogram selection unit 190. Details will be described in S300 and the like.

(Presentation sound control unit 220)
The presented sound control unit 220 determines a plurality of sounds having different frequencies and sound pressures from sounds stored in the presented sound storage unit 111 in order to measure the loudness of the user 100. The presentation sound control unit 220 determines the number of presentations of the determined sound and the presentation order. The presenting sound control unit 220 transmits information on a plurality of presenting sounds, the order of presenting the presenting sounds, and the interval for presenting the presenting sounds to the electroacoustic transducer 130.

(EEG selection unit 190)
The electroencephalogram selection unit 190 receives the typical pattern selected by the typical pattern selection unit 180. The electroencephalogram selection unit 190 expands and contracts the received typical pattern so as to be similar to the measured electroencephalogram in the expansion / contraction range and the time range held in advance. The latency of the N1 component in the waveform when the typical pattern is expanded and contracted so as to be most similar to the measured brain wave is acquired. The electroencephalogram selection unit 190 transmits the acquired latency of the N1 component to the latency estimation unit 200. Details will be described in S400 and the like.

(Latency estimation unit 200)
The latency estimation unit 200 receives the latency of the N1 component acquired from the electroencephalogram selection unit 190. The latency estimation unit 200 estimates the latency of the N1 component for each frequency of the presentation sound and the sound pressure of the presentation sound. The latency estimation unit 200 transmits the N1 component latency for each frequency of the presentation sound and the sound pressure of the presentation sound to the measurement completion determination unit 210.

(Measurement completion determination unit 210)
The measurement completion determination unit 210 receives from the latency estimation unit 200 the latency of the N1 component for each frequency of the presentation sound and the sound pressure of the presentation sound. The measurement completion determination unit 210 determines measurement completion when the number of received N1 component latencies is equal to or greater than a predetermined number. The measurement completion determination unit 210 obtains a loudness increase curve for each frequency with reference to a predetermined relationship between the sound pressure of the presenting sound and the latency of the N1 component. Details will be described later.

  FIG. 6 is a flowchart showing the operation of the loudness measurement system 10. An outline of the operation of the loudness measurement system 10 will be described with reference to FIG.

(S100)
The loudness measurement system 10 starts auditory measurement after receiving an input to start measurement. For example, the presenting sound determination unit 110 receives an input for starting measurement.

(S200)
The presentation sound determination unit 110 determines a plurality of presentation sounds from the sounds stored in the presentation sound storage unit 111 with reference to the loudness correspondence relationship between the sound pressure value and the loudness value. In order to be used for selecting a typical pattern, the presentation sound determined by the presentation sound determination unit 110 is also referred to as “initial presentation sound”. The determined initial presentation sounds have different frequencies and sound pressures having the same loudness value. The presenting sound generation unit 120 generates information on the initial presenting sound including the order of presenting the presenting sound and the interval of presenting the presenting sound. The electroacoustic transducer 130 presents a presentation sound to the user 100 based on the generated information of the initial presentation sound. The electroencephalogram measurement unit 150 measures the electroencephalogram of the user 100 when the initial presentation sound is presented.

(S300)
The sample pattern generation unit 160 generates an electroencephalogram sample pattern by averaging the brain waves with respect to the initial presentation sound. The generated brain wave sample pattern is obtained by averaging the brain waves for a plurality of sounds having different frequencies. The typical pattern selection unit 180 expands / contracts the typical patterns of the plurality of brain waves including the N1 component and the P2 component stored in the typical pattern storage unit 170 within a predetermined expansion / contraction range and time range. The typical pattern selection unit 180 compares the stretched typical pattern with the generated electroencephalogram sample pattern, and selects the typical pattern closest to the electroencephalogram sample pattern.

(S400)
The presented sound control unit 220 determines a plurality of presented sounds having different frequencies and sound pressures. The electroacoustic transducer 130 presents the user 100 with a plurality of presenting sounds having different frequencies and sound pressures. The electroencephalogram measurement unit 150 measures the electroencephalogram of the user 100 when the presentation sound is presented.
The electroencephalogram selection unit 190 expands / contracts the typical pattern selected in S300 within a predetermined expansion / contraction range and time range, and compares it with the measured electroencephalogram. The latency of the N1 component of the typical pattern expanded and contracted so as to be closest to the measured electroencephalogram is acquired. The latency estimation unit 200 uses the latency of the N1 component acquired by the electroencephalogram selection unit 190 to estimate the latency of the N1 component for each frequency of the presentation sound and the sound pressure of the presentation sound. The measurement completion determination unit 210 refers to a predetermined relationship between the sound pressure of the presenting sound and the latency of the N1 component, and for each frequency of the presenting sound, the estimated N1 component latent for the sound pressure of the presenting sound. Find the loudness increase curve corresponding to the change of time. Unlike the process of S200, the measurement completion determination unit 210 uses an electroencephalogram for each frequency of the presentation sound.

(S500)
The measurement completion determination unit 210 outputs a loudness increase curve for each frequency and ends the measurement.

  In S300, a typical pattern is selected while expanding / contracting the typical pattern, and a loudness curve is obtained using the typical pattern. As a result, it is possible to generate a loudness curve robustly with respect to the time series data of the electroencephalogram. Since the characteristics of the measured electroencephalogram differ depending on the individual difference in potential or the electrode mounting position, the amplitude or latency value of the event-related potential including the more accurate N1 component can be measured by using the typical pattern. .

  Hereinafter, each step of the flowchart shown in FIG. 6 will be described in detail.

  FIG. 7 is a flowchart showing details of step S200 in the flowchart shown in FIG.

(S201)
The presentation sound determination unit 110 selects a plurality of sounds having different frequencies from the sounds stored in the presentation sound storage unit 111. The presented sound determination unit 110 refers to the previously stored loudness correspondence, and determines sound pressures that have the same loudness value for a plurality of determined presented sounds. That is, the presentation sound determination unit 110 determines the information of the presentation sound including the frequency and sound pressure information for each of the plurality of presentation sounds.

  A human can hear a sound having a frequency in the range of 1 kHz to 3 kHz with high sensitivity. On the other hand, it is known that a human being has a minimum audible value and a loudness different depending on the frequency with respect to a sound having a frequency lower than 1 kHz or a sound having a frequency higher than 3 kHz. Therefore, for example, each frequency is determined based on the sound pressure at which a normal hearing person becomes 70 dBSPL for a sound of 1 kHz. For example, for a pure sound of 500 Hz, 1 kHz, and 2 kHz, a sound pressure value that is 90 dBSL is determined.

(S202)
The presentation sound generation unit 120 determines the order in which the presentation sounds determined in S201 are presented and the interval at which the presentation sounds are presented. For example, the presentation sound generation unit 120 holds in advance information on the number of presentation sounds and the presentation interval of the presentation sounds. The presentation sound generation unit 120 refers to the number of presentations and the presentation interval, and determines the order in which the presentation sounds are presented and the interval at which the presentation sounds are presented. The presenting sound generation unit 120 determines the order of presenting so that presenting sounds having the same frequency are not presented continuously.

  The presenting sound generation unit 120 receives information on the number of times the presenting sound is presented and the interval at which the presenting sound is presented, and determines the order in which the presenting sounds are presented and the presenting sound presenting interval by using the received information. Also good. For example, the presentation interval is 800 ms. Alternatively, the presentation interval is any time selected from the range of 400 ms to 1200 ms. The number of presentations is, for example, 3 times.

  The electroacoustic transducer 130 converts the presentation sound into an acoustic signal using the frequency of the presentation sound, the information on the sound pressure of the presentation sound, the order in which the presentation sounds are presented, and the interval at which the presentation sounds are presented. Present.

(S203)
The electroencephalogram measurement unit 150 measures the electroencephalogram of the user 100 using a plurality of electrodes 140 attached on the scalp of the user 100. In FIG. 26, the electrode position of the international 10-20 method (10-20 System) is shown. FIG. 26 is a diagram seen from above the human head. For example, the measurement electrode 142 is attached to Fz, Cz, Pz, the right eye, the right eye, the left mastoid, and the right mastoid shown in FIG. 26, and the reference electrode 141 is attached to the nose. A “mastoid” is the mastoid process of the skull at the bottom of the base of the back of the ear. The electroencephalogram measurement unit 150 measures an electroencephalogram corresponding to the potential difference between the reference electrode 141 and the measurement electrode 142.

(S204)
The sample pattern generation unit 160 cuts out an electroencephalogram having a certain time width as the starting point of the presented time from the electroencephalogram measured by the electroencephalogram measurement unit 150 for each presentation sound. For example, an electroencephalogram in the range of −100 ms to 400 ms is cut out from the presentation time of the presentation sound. “−100 ms” means 100 ms before the time when the presentation sound is presented. The sample pattern generation unit 160 averages the brain waves cut out for each presentation sound to generate a sample pattern of brain waves. That is, the sample pattern generation unit 160 generates an electroencephalogram sample pattern obtained by adding and averaging all the electroencephalograms for the initial presentation sounds having different frequencies and the same loudness.

(S205)
The typical pattern selection unit 180 determines whether the N1 component and the P2 component can be detected in the sample pattern of the electroencephalogram generated in step S204. The typical pattern selection unit 180 detects a negative peak greater than or equal to the first threshold after a range from 80 ms to 130 ms starting from the time when the presentation sound was presented. The detected negative peak is an N1 component candidate. The typical pattern selection unit 180 detects a positive peak greater than or equal to the second threshold after a range from 150 ms to 280 ms starting from the time when the presentation sound was presented. The detected positive peak is a P2 component candidate.

  When both the N1 component candidate and the P2 component candidate can be detected, the typical pattern selection unit 180 determines that the N1 component and the P2 component can be detected from the electroencephalogram sample pattern. At this time, the process flow proceeds to step S300. When at least one of the N1 component candidate and the P2 component candidate cannot be detected, the typical pattern selection unit 180 determines that the N1 component and the P2 component cannot be detected from the electroencephalogram sample pattern. At this time, the processing flow returns to step 201. When returning to step 201, the presentation sound determination unit 110 determines a presentation sound having a loudness greater than the previously determined presentation sound.

  FIG. 8 is a flowchart showing details of step S300 in the flowchart shown in FIG. The typical pattern selection unit 180 performs the process of step S300. First, the typical pattern selection unit 180 will be described. FIG. 4 is a detailed block diagram of the typical pattern selection unit 180. The typical pattern selection unit 180 includes a detection unit 181, a latency difference calculation unit 182, and a sample pattern matching unit 183. Each configuration will be described below.

(Detector 181)
The detection unit 181 detects an N1 component candidate and a P2 component candidate in an electroencephalogram sample pattern. Starting from the time when the presenting sound was presented, a negative peak greater than or equal to the first threshold value is detected after a range of between 80 ms and 130 ms. The detected negative peak is an N1 component candidate. The typical pattern selection unit 180 detects a positive peak greater than or equal to the second threshold after a range from 150 ms to 280 ms starting from the time when the presentation sound was presented. The detected peak that has been detected is a P2 component candidate. The detection unit 181 acquires the latency or amplitude of the N1 component candidate and the P2 component candidate.

(Latency acquisition unit 182)
The latency acquisition unit 182 acquires the difference between the latency of the P2 component and the latency of the N1 component, the time of the N1 component, and the time of the P2 component.

(Typical pattern matching unit 183)
The typical pattern matching unit 183 determines the expansion / contraction range using a predetermined magnification and the difference between the latency of the P2 component and the latency of the N1 component. The typical pattern matching unit 183 determines a time range using a predetermined time range and the difference between the P2 component latency and the N1 component latency, or the N1 component time and the P2 component time.

  The typical pattern matching unit 183 expands / contracts the typical pattern using the determined expansion / contraction range and time range, and selects an expanded / contracted typical pattern similar to the sample pattern of the electroencephalogram.

  Next, the process of step S300 shown in FIG. 8 will be described.

(S301)
The detection unit 181 determines the N1 component and the P2 component using the N1 component candidate and the P2 component candidate detected in step S205. When the detection unit 181 detects one N1 component candidate, the N1 component candidate is determined as the N1 component. When the detection unit 181 detects one P2 component candidate, the P2 component candidate is determined as the P2 component.

  When the detection unit 181 detects a plurality of N1 component candidates, one N1 component is determined from them. When the detection unit 181 detects a plurality of P2 component candidates, one P2 component is determined from them. For example, the detection unit 181 determines the N1 component candidate that is a negative peak having the smallest potential value as N1. The negative peak with the smallest potential value is synonymous with the negative peak with the largest amplitude. Alternatively, the detection 181 acquires the latency of the N1 component candidate. The detection unit 181 determines the N1 component candidate having the latency closest to 100 ms of the standard N1 latency held in advance as the N1 component.

  The detection unit 181 determines the P2 component candidate corresponding to the positive peak having the largest potential value after the range of 80 ms or more and 130 ms or less from the determined latency of the N1 component as the P2 component. The positive peak having the largest potential value is synonymous with the positive peak having the largest amplitude.

(S302)
The latency acquisition unit 182 acquires a latency difference that is a difference between the latency of the P2 component and the latency of the N1 component. The latency acquisition unit 182 acquires the time of the N1 component and the time of the P2 component. The time of the N1 component corresponds to the elapsed time from the time when the presentation sound is presented to the N1 component. The time of the P2 component corresponds to the elapsed time from the time when the presentation sound is presented to the P2 component.

(S303)
The typical pattern storage unit 170 stores a typical pattern of an electroencephalogram having an N1 component and a P2 component. For example, the typical pattern of the electroencephalogram is information on the potential value at each time from the stimulus.

  The typical pattern matching unit 183 determines the expansion / contraction range of the N1 component and the P2 component included in the typical pattern used in S305 described later.

  For example, the typical pattern matching unit 183 holds a predetermined magnification. The typical pattern matching unit 183 may determine a predetermined magnification as the expansion / contraction range. Alternatively, the typical pattern matching unit 183 determines the expansion / contraction range based on the product of the latency difference acquired in S302 and a predetermined magnification. When the latency difference is 100 ms and the predetermined magnification is 0.8 times or more and 1.2 times or less, the typical pattern matching unit 183 determines an expansion / contraction range in which the latency difference is 80 ms or more and 120 ms or less.

  The expansion / contraction range may be determined for each typical pattern. When all the typical patterns are standardized on the time axis so that the latency is the same, an expansion / contraction range common to all the typical patterns is determined. For example, the standardized typical pattern has a difference of 100 ms between the latency of the N1 component and the latency of the P2 component included in all the typical patterns.

(S304)
The typical pattern matching unit 183 determines a time range for matching the sample pattern of the electroencephalogram and the typical pattern used in S305 described later. The typical pattern matching unit 183 determines the time range using the time of the N1 component and the time of the P2 component of the sample pattern of the electroencephalogram and information of a predetermined time width.

  FIG. 27 shows an example of an electroencephalogram sample pattern. In FIG. 27, the vertical axis represents the potential corresponding to the electroencephalogram, and the horizontal axis represents time. In FIG. 27, the time from the time having a brain wave potential of 0 μV to the time having the N1 component is represented as T1. T1 is referred to as a rise time. In FIG. 27, the time from the time having the P2 component to the time having the electroencephalogram potential of 0 μV is represented as T2. T2 is referred to as a fall time.

  The characteristic of the N1 component also appears at the rise time before the time of the N1 component. The characteristic of the P2 component also appears at the fall time after the time of the P2 component. Therefore, the typical pattern matching unit 183 determines a time range in which the rising time and the falling time are added to the time of the P2 component from the time of the N1 component as a matching time range.

  Consider a case where the time of the N1 component is 100 ms and the time of the P2 component is 200 ms. For example, the typical pattern matching unit 183 holds time width information in which the rise time is 50 ms and the fall time is 80 ms. 50 ms obtained by subtracting the fall time from the time of the N1 component is the lower limit of the time range. The upper limit of the time range is 280 ms obtained by adding the fall time from the time of the P2 component. That is, the typical pattern matching unit 183 determines a time range from 50 ms to 280 ms. Alternatively, the typical pattern matching unit 183 holds the rise time as 0.4 times the latency of the N1 component and the fall time as 0.6 times the latency difference between N1 and P2. 40 ms obtained by subtracting 0.4 times the latency of the N1 component from the time of the N1 component is the lower limit of the time range. The upper limit of the time range is 320 ms obtained by adding 0.6 times the latency of the P2 component from the time of the P2 component. That is, the typical pattern matching unit 183 determines a time range from 40 ms to 320 ms.

(S305)
The typical pattern matching unit 183 obtains the similarity with the sample pattern of the electroencephalogram generated in S204 from the typical pattern stored in the typical pattern storage unit 170.

  Specifically, the typical pattern matching unit 183 matches the typical pattern and the EEG sample pattern using the expansion / contraction range determined in S303 and the time range determined in S304.

  Specifically, the correlation coefficient between the electroencephalogram sample pattern and the typical pattern is obtained in the time range determined in step S304. At this time, the correlation coefficient between the electroencephalogram sample pattern and the typical pattern is obtained by expanding / contracting the typical pattern within the expansion / contraction range determined in S303.

  FIG. 10 schematically shows a matching procedure. The vertical axis in FIG. 10 is the potential corresponding to the electroencephalogram, and the horizontal axis is time.

  The typical pattern is stretched within the determined stretch range. Here, the expansion / contraction range is 0.9 to 1.1 times. As shown in FIG. 10, a typical pattern 1000 stretched 0.9 times, a typical pattern 1003 stretched 1.0 times, and a typical pattern 1004 stretched 1.1 times are prepared.

  Next, in the determined time range (T3), a correlation coefficient between the electroencephalogram sample pattern and the typical pattern expanded and contracted in S1001 is obtained.

  A typical pattern 1000 shown in FIG. 10 is a waveform having a time shorter than the time range (T3). In this case, the correlation coefficient between the sample pattern of the electroencephalogram and the typical pattern expanded and contracted in S1001 is obtained while shifting the typical pattern 1000 expanded and contracted 0.9 times in the time direction. At time T4 when the typical pattern 1000 is included, a correlation coefficient between the typical pattern 1000 and the sample pattern of the electroencephalogram is obtained. Next, the typical pattern 1000 expanded and contracted 0.9 times is shifted to a typical pattern 1001 expanded and contracted 0.9 times. At a time T5 when the typical pattern 1001 is included, a correlation coefficient between the typical pattern 1001 and the brain wave sample pattern is obtained. Next, the typical pattern 1001 stretched 0.9 times is shifted to a typical pattern 1002 stretched 0.9 times. At time T6 when the typical pattern 1002 is included, a correlation coefficient between the typical pattern 1002 and the sample pattern of the electroencephalogram is obtained. For example, the highest correlation coefficient obtained by shifting the time direction by 10 ms is set as a correlation coefficient between a typical pattern expanded and contracted 0.9 times and a sample pattern of an electroencephalogram. The time ranges T4, T5, and T6 are included in the determined time range (T3).

  A typical pattern having a time width longer than the determined time range (T3) may be stored in the typical pattern storage unit 170. In this case, the correlation coefficient between the typical pattern and the EEG sample pattern is obtained in the determined time range (T3).

(S306)
The typical pattern matching unit 183 refers to the correlation coefficient obtained in S305 and selects a typical pattern having the highest correlation coefficient.

(S400)
FIG. 11 is a flowchart showing details of step S400 in the flowchart shown in FIG. The electroencephalogram data selection unit 190 mainly performs the process of step S400. First, the electroencephalogram data selection unit 190 will be described. FIG. 5 is a block diagram of the electroencephalogram data selection unit 190. The electroencephalogram data selection unit 190 includes a pattern expansion / contraction unit 191, an electroencephalogram extraction unit 192, a pattern matching unit 193, and a selection determination unit 194.

(Pattern expansion / contraction part 191)
The pattern expansion / contraction part 191 expands / contracts the typical pattern in a predetermined expansion / contraction range and time range.

(Electroencephalogram clipping unit 192)
The electroencephalogram clipping unit 192 receives the time when the presentation sound is presented from the electroacoustic transducer 130. The electroencephalogram extraction unit 192 extracts an electroencephalogram having a predetermined time width from the time point when the presentation sound is presented, from the measured electroencephalogram.

(Pattern matching unit 193)
The pattern matching unit 193 obtains a correlation function between the measured pattern and the typical pattern expanded and contracted by the pattern expansion / contraction unit 191 for each condition of the expansion / contraction range and the time range.

(Selection determination unit 194)
The selection determining unit 194 specifies the expansion / contraction range and the time range of the typical pattern that maximizes the correlation function obtained by the pattern matching unit 193. The latency of the N1 component of the typical pattern in the specified expansion / contraction range and time range is acquired.

  Next, the process of step S400 shown in FIG. 11 will be described.

(S411)
The presentation sound control unit 220 determines a plurality of presentation sounds having different frequencies. Unlike S201, the presentation sound control unit 220 determines a presentation sound having a plurality of different sound pressures for each determined frequency.

(S412)
The presentation sound generation unit 120 generates the determined presentation sound. The electroacoustic transducer 130 converts the presentation sound generated by the presentation sound generation unit 120 into an acoustic signal and presents it to the user 100.

(S413)
The electroencephalogram measurement unit 150 measures an electroencephalogram corresponding to the potential difference between the plurality of electrodes 140 attached to the user 100. The measured electroencephalogram includes an electroencephalogram at the time when the presentation sound is presented.

(S414)
The electroencephalogram extraction unit 192 extracts an electroencephalogram having a predetermined time width from the electroencephalogram measured in S413 as a starting point. For example, the electroencephalogram extraction unit 192 extracts an electroencephalogram in a section of −100 ms or more and +400 ms or less using the presentation start point of the presentation sound.

(S415)
The pattern expansion / contraction unit 191 expands / contracts the typical pattern selected in step S300 within a predetermined expansion / contraction range. For example, the pattern expansion / contraction part 191 expands / contracts the typical pattern in the range of 0.5 to 1.5 times. The pattern matching unit 193 matches the typical pattern expanded / contracted by the pattern expansion / contraction unit 191 and the electroencephalogram cut out in S414 within a predetermined matching range. The predetermined matching time range is, for example, a range from 70 ms to 450 ms starting from the time when the presentation sound is presented.

  As in S300, the pattern matching unit 193 obtains a correlation coefficient between the electroencephalogram measured in S413 and the typical pattern in the time range determined in step S414, as shown in FIG. The pattern matching unit 193 obtains a correlation coefficient between the electroencephalogram measured in S413 and the typical pattern for each predetermined expansion / contraction range and time range condition.

(S416)
The pattern matching unit 193 determines whether the maximum value among the correlation coefficients obtained in step S415 is equal to or greater than a reference value. This standard indicates the lower limit of the degree of similarity between the measured electroencephalogram and the typical pattern. For example, the reference is 0.5. If the maximum correlation coefficient is greater than or equal to the reference, the process proceeds to step S417. If the maximum correlation coefficient is less than the reference, it is determined that the brain wave is not adopted as data, and the process proceeds to step S418.

(S417)
The selection determination unit 194 specifies the expansion / contraction range and the time range of the typical pattern that maximizes the correlation coefficient obtained in S415. The selection determination unit 194 acquires the latency of the N1 component in the typical pattern when the correlation coefficient is maximized. The selection determination unit 194 determines the latency of the N1 component acquired using the typical pattern as the latency of the N1 component of the electroencephalogram measured in S413.

(S418)
The measurement completion determination unit 210 determines whether or not a predetermined number or more of N1 components of the electroencephalogram have been determined. The measurement completion determination unit 210 may hold a predetermined number for each frequency of the presentation sound. For example, the measurement completion determination unit 210 holds 5 as a predetermined number.

  When the measurement completion determination unit 210 determines that the N1 components of the brain waves equal to or greater than the predetermined number have been determined, the process proceeds to step S420. When the measurement completion determination unit 210 determines that N1 components of the brain waves equal to or greater than the predetermined number have not been determined, the process returns to step S411.

(S421)
The latency estimation unit 200 obtains the average latency of the N1 component for each frequency and sound pressure of the presented sound.

(S422)
The measurement completion determination unit 210 summarizes the latency of the N1 component for each sound pressure of the presentation sound for each frequency of the presentation sound, and generates a loudness increase curve for each frequency.

  Using the method described below, measurement completion determination unit 210 generates a loudness increase curve. The measurement completion determination unit 210 holds in advance the relationship between the sound pressure of the presentation sound and the latency of the N1 component, and the loudness. The measurement completion determination unit 210 generates a loudness increase curve with reference to the relationship between the sound pressure of the presented sound and the latency of the N1 component and the loudness.

  FIG. 9A shows the relationship between the normal hearing person as an example of the relationship between the sound pressure of the presented sound and the latency of the N1 component and the loudness. In FIG. 9A, the vertical axis represents the latency (ms), and the horizontal axis represents the sound pressure (dBHL) of the presented sound. The shaded portion shown in FIG. 9A shows the relationship between the sound pressure of the presenting sound possessed by a normal hearing person and the latency of the N1 component. Although not shown in FIG. 9A, the loudness value of the normal hearing person is held for each sound pressure of the presenting sound and the latency of the N1 component. The measurement completion determination unit 210 may have a relationship between the sound pressure of the presented sound at the boundary portion indicated by the dotted line and the latency of the N1 component and the loudness in the shaded portion information.

  The white circles shown in FIG. 9A are values corresponding to the sound pressure of the presentation sound and the latency of the N1 component obtained in S421. For the presentation sound having a sound pressure of 70 dBHL or more, the dispersion of the latency of the N1 component is small. For example, the loudness increase curve is generated using the latency of the N1 component with respect to a sound pressure of 70 dBHL or more with small variance. You may obtain | require in the whole range of the sound pressure of the measured presentation sound.

  The measurement completion determination unit 210 refers to the relationship between the sound pressure of the presentation sound and the latency of the N1 component, and obtains the difference between the sound pressure of the presentation sound and the latency of the N1 component obtained in S421. FIG. 9A shows a point of the N1 component having a latency of about 120 ms for a presentation sound having a sound pressure of 70 dBHL. A normal hearing person has an N1 component latency of about 120 ms for a presentation sound having a sound pressure of 45 dBHL. The latency of the N1 component of the user 100 for a presentation sound having a sound pressure of 70 dBHL is the same as the N1 component of a normal hearing person for a presentation sound having a sound pressure of 45 dBHL. Therefore, the loudness value of the normal hearing person corresponding to the sound pressure of 45 dBHL is the value of the loudness of the user 100 with respect to the presentation sound having the sound pressure of 70 dBHL. As shown in FIG. 9A, the latency of the N1 component of the user 100 for the presentation sound having a sound pressure of 80 dBHL corresponds to the latency of the N1 component of the normal hearing person for the presentation sound having a sound pressure of 65 dBHL. Further, the latency of the N1 component of the user 100 with respect to the presentation sound having a sound pressure of 90 dBHL corresponds to the latency of the N1 component of the normal hearing person with respect to the presentation sound having a sound pressure of 90 dBHL.

  FIG. 9B shows a loudness increase curve generated based on these results. In FIG. 9B, the vertical axis is loudness, and the horizontal axis is sound pressure (dBHL). The solid line shown in FIG. 9B is a straight line connecting the loudness values of the user 100 with respect to the presentation sound having 70 dBHL, 80 dBHL, and 90 dBHL. This solid line is the generated loudness increase curve. A dotted line shown in FIG. 9B is a loudness increase curve of a normal hearing person. As shown in FIG. 9B, the loudness increase curve may not be a curve, or may be a table in which the sound pressure of the presentation sound and the loudness value are associated with each other.

  In the present embodiment, the typical pattern matching unit 183 may select a plurality of typical patterns, and the pattern matching unit 193 may match the measured electroencephalogram using the selected typical patterns. For example, the typical pattern matching unit 183 can select a plurality of typical patterns having a degree of similarity higher than a predetermined threshold with respect to the measured electroencephalogram. It is also possible to select the top few typical patterns having high similarity. The pattern matching unit 193 may match the measured electroencephalogram using a plurality of selected typical patterns. Since the absolute value of the potential of the electroencephalogram is small, it is difficult to measure the electroencephalogram, and in the case of one typical pattern selected by the typical pattern matching unit 183, there is a possibility of selecting an incorrect pattern. By selecting a plurality of typical patterns, the typical pattern matching unit 183 can change to another typical pattern even when an incorrect pattern is selected.

(Modification of Embodiment 1)
FIG. 12 shows a configuration of a hearing aid adjustment system including the loudness measurement system according to the first embodiment. The loudness measurement system shown in FIG. 13 includes an HTL input unit 320, a hearing aid input / output characteristic setting unit 330, and a hearing aid 340 in addition to the auditory measurement device of the first embodiment. The loudness measurement system shown in FIG. Compared to the loudness measurement system of the first embodiment, the configuration is the same except that the presentation sound determination unit 110 is replaced with the presentation sound determination unit 310 and the presentation sound control unit 220 is replaced with the presentation sound control unit 350. The same components as those in the first embodiment are denoted by the same reference numerals, and description thereof is omitted as appropriate. Hereinafter, a configuration different from the auditory measurement apparatus according to the first embodiment will be described. Each configuration of the hearing aid adjustment system is connected by wire or wirelessly.

(HTL input unit 320)
The HTL input unit 320 inputs the minimum audible value (Hearing Threshold Level: HTL) for each frequency of the user 100 to the presentation sound determination unit 310 and the presentation sound control unit 350.

Note that the HTL input unit 320 may hold an HTL for each frequency of the user 100 in advance, or may receive an HTL for each frequency of the user 100 from the outside. The HTL input unit 320 includes, for example, a numeric keypad and a cursor key.
(Presentation sound determination unit 310)
The presentation sound determination unit 310 refers to the pure tone stored in the presentation sound storage unit 111 and determines an initial presentation sound that is a pure tone having a different frequency.

  The presentation sound determination unit 310 receives the HTL from the HTL input unit 320. The presentation sound determination unit 310 holds the relationship between the sound pressures having the same loudness value as the HTL held in advance. For example, FIG. 14 shows a relationship between sound pressures having the same loudness value as that of HTL held in advance. FIG. 14 shows the relationship between HTL (dBSPL) and the corresponding loudness value. The presented sound determination unit 310 determines a sound pressure at which the loudness value corresponding to the received HTL is the same. The sound pressure corresponding to the determined loudness value is determined for each frequency with reference to the loudness correspondence. Details will be described later.

  That is, the presentation sound determination unit 310 refers to the received HTL and the relationship between the HTL and the sound pressure held in advance, and determines an initial presentation sound having a sound pressure that is different in frequency and has the same loudness. . The presentation sound determination unit 310 transmits information on a plurality of presentation sounds to the presentation sound generation unit 120. The “presentation sound information” includes information on the frequency and sound pressure of the presentation sound.

(Presentation sound control unit 350)
The presentation sound control unit 350 receives the HTL from the HTL input unit 320. In order to measure the loudness of the user 100, the presentation sound control unit 350 determines a plurality of sounds having different frequencies and sound pressures from the sound stored in the presentation sound storage unit 111 with reference to the received HTL. To do. The presentation sound control unit 220 determines the number of presentations of the determined sound and the presentation order. The presenting sound control unit 220 transmits information on a plurality of presenting sounds, the order of presenting the presenting sounds, and the interval for presenting the presenting sounds to the electroacoustic transducer 130.

(Characteristic setting unit 330)
The characteristic setting unit 330 maintains a predetermined relationship between the loudness and sound pressure of a normal hearing person. The characteristic setting unit 330 receives the loudness increase curve from the measurement completion determination unit 210. The characteristic setting unit 330 sets a characteristic corresponding to a difference between a predetermined relationship between the loudness and sound pressure of a normal hearing person and a loudness increase curve. The characteristic setting unit 330 transmits the characteristic set to the hearing aid 340.

(Hearing aid 340)
The hearing aid 340 sets the characteristic based on the data received from the characteristic setting unit 330.
The hearing aid 340 receives sound from the outside, adjusts the received sound based on the set characteristics, and presents it to the user 100.

FIG. 13 shows a flowchart of processing of the hearing aid adjustment system. Hereinafter, the same steps are performed in steps having the same steps as those in the flowchart of the processing of the loudness measuring apparatus shown in FIG.
(S200)
The HTL input unit 210 inputs the HTL for each frequency to the presentation sound determination unit 301. For example, HTL is input every 250 Hz, 500 Hz, 1000 Hz, 2000 Hz, and 4000 Hz. The presented sound determination unit 301 determines the sound pressure corresponding to the received HTL with reference to the relationship of the sound pressure having the same loudness value as that of the HTL. The presentation sound determination unit 301 refers to the received frequency and the determined sound pressure, and selects a plurality of sounds having different frequencies from the sounds stored in the presentation sound storage unit 111.

  The presented sound determination unit 301 selects a plurality of sounds having different frequencies from each other with reference to the received HTL, and then selects the sound by referring to the relationship between sound pressures having the same loudness value as the HTL. The sound pressure of the sound may be determined.

  The present sound generation unit 120, the electroacoustic transducer 130, and the electroencephalogram measurement unit 150 perform the same processing as the loudness measurement system of the first embodiment, and measure the user's electroencephalogram when the initial present sound is presented.

(S300)
In S300, the hearing aid adjustment system performs the same processing as the loudness measurement system shown in FIG. 6, and selects a typical pattern closest to the sample pattern of the electroencephalogram.

(S401)
The presentation sound control unit 350 receives the HTL from the HTL input unit 320. The presented sound control unit 350 refers to a predetermined relationship between HTL and sound pressure, and determines a plurality of presented sounds having different frequencies and sound pressures.

  The acoustic transducer 130, the electroencephalogram measurement unit 150, the electroencephalogram selection unit 190, the latency estimation unit 200, and the measurement completion determination unit 210 perform the same processing as in the loudness measurement system of Embodiment 1, and output a loudness increase curve for each frequency. To do.

(S500)
The characteristic setting unit 330 sets a characteristic corresponding to a difference from the output loudness increase curve with reference to a predetermined relationship between the loudness and sound pressure of the normal hearing person. FIG. 15 shows a relationship 1500 between the normal hearing person's loudness and sound pressure, and a relationship 1501 between the hearing person's loudness and sound pressure.
The output loudness increase curve corresponds to the relationship 1501 between the loudness and sound pressure of the hearing impaired. For example, the characteristic setting unit 330 sets a characteristic corresponding to the difference between the loudness increase curve 1500 for the normal hearing ear and the loudness increase curve 1501 for the deaf ear shown in FIG. When the difference between the loudness of the hearing-impaired ear and the loudness of the normal-hearing ear is equal to or greater than a predetermined value, a large gain indicated by a white arrow shown in FIG. If the loudness of the deaf ear is less than a predetermined value than the loudness of the normal ear, a small gain is given or no gain is given. Thereby, it adjusts so that it may enter into the audible range of a hearing-impaired ear with respect to the sound pressure of the audible range of a normal hearing ear. The characteristic setting unit 330 transmits the set characteristic to the hearing aid 340. The hearing aid 340 sets the characteristic of the hearing aid 340 based on the data received from the characteristic setting unit 330.

  In the first embodiment and the modification of the first embodiment, the matching is performed by comparing the correlation coefficient by the region shift. However, DP matching may be performed using the amount of change in potential as an index. However, in the case of DP matching, a limit is set for expansion and contraction in the time direction during matching to prevent nonlinear deformation of the time pattern.

(Modification 2 of Embodiment 1)
FIG. 16 shows a part of the loudness measurement system 12 according to the second modification of the first embodiment. The loudness measurement system 12 of the second modification of the first embodiment is similar to the loudness measurement system 10 of the first embodiment except that the latency acquisition unit 182 included in the typical pattern selection unit 180 of the first embodiment is not included. It is the same. Each configuration of the loudness measurement system 12 is connected by wire or wireless.

  In step S303 shown in FIG. 8, the typical pattern matching unit 183 linearly expands and contracts the typical pattern to match the sample pattern of the electroencephalogram. Moreover, in step S415 shown in FIG. 11, the pattern expansion / contraction part 191 linearly expands / contracts a typical pattern, and matches with the cut-out brain wave.

  On the other hand, the typical pattern matching unit 183 and the pattern expansion / contraction unit 191 of the present modification expand / contract the typical pattern nonlinearly. The typical pattern matching unit 183 and the pattern expansion / contraction unit 191 of the present modification have a small expansion / contraction rate near the zero point on the time axis, which is the stimulus presentation time point, and a large expansion / contraction rate at a point far from the zero point. Stretch non-linearly. In the electroencephalogram, the shorter the latency component, the smaller the latency change due to the stimulation condition, and the longer the latency component, the greater the latency variation due to the stimulation condition. Therefore, the nonlinear expansion / contraction makes it easy to match the typical pattern and the brain wave sample pattern or the extracted brain wave, and the estimation error is reduced.

  The loudness measurement system according to the second modification has the same processing as that shown in FIG. Processing different from the processing of the loudness measurement system according to the first embodiment will be described. 17 and 19 show a flowchart of processing of the loudness measurement system according to the second modification of the first embodiment.

(S301)
The detection unit 181 determines the N1 component and the P2 component of the sample pattern of the electroencephalogram. In addition, the detection unit 181 acquires the time of the N1 component and the time of the P2 component.

(S1303)
For each of the typical patterns stored in the typical pattern storage unit 170, the typical pattern matching unit 183 determines the range of the expansion / contraction function parameter used when expanding / contracting the typical pattern on the time axis, which is used in S305 described later.

  FIG. 18 shows an example of the time position before expansion / contraction and the time position after expansion / contraction. A curve indicated by a solid line in FIG. 18 is an example of a function that gives the maximum waveform expansion, and is, for example, a quadratic curve. A curve indicated by a broken line in FIG. 18 is an example of a function that gives the minimum expansion / contraction of the waveform, and is a quadratic curve having a smaller coefficient than the solid line.

  As shown in FIG. 18, a function that gives maximum extension is applied to a time interval of 100 ms to 300 ms, that is, a 200 ms interval, starting from a reference point on the time axis of a typical pattern. As a result, 100 ms is converted to 10 ms, and 300 ms is converted to 360 ms. The time length of the time section is 350 ms (360 ms-10 ms). A function that gives a minimum reduction is applied to a time interval of 100 ms to 300 ms. As a result, 100 ms is converted to 20 ms, and 300 ms is converted to 180 ms. The time length of the time interval is reduced to 160 ms (180 ms-20 ms).

  The detection unit 181 has the latency of the N1 component of the typical pattern smaller than the latency of the N1 component of the EEG sample pattern acquired in step S301, and the latency of the P2 component of the typical pattern is the P2 component of the EEG sample pattern. The parameter of the function giving the maximum extension is determined so as to be larger than the latency. In addition, the minimum reduction is performed so that the latency of the N1 component of the typical pattern is larger than the latency of the N1 component of the sample pattern of the electroencephalogram and the latency of the P2 component of the typical pattern is smaller than the P2 latency of the sample pattern of the electroencephalogram Determine the parameters of the function that gives The function parameter is, for example, a coefficient of a quadratic expression.

(S304)
Similar to S304 shown in FIG. 8, the typical pattern matching unit 183 determines the time range using the time of the N1 component and the time of the P2 component of the sample pattern of the electroencephalogram and the information of the predetermined time width.

(S1305)
The typical pattern matching unit 183 obtains a correlation coefficient between the sample pattern of the electroencephalogram and the typical pattern in the time range determined in step S304. At this time, the correlation coefficient between the electroencephalogram sample pattern and the typical pattern is obtained by expanding and contracting the typical pattern using the parameter of the function determined in S1303.

(S1306)
The typical pattern matching unit 183 refers to the correlation coefficient obtained in S305 and selects a typical pattern having the highest correlation coefficient. The typical pattern matching unit 183 outputs, to the electroencephalogram selection unit 190, the parameter of the function that gives the maximum expansion and the parameter of the function that gives the minimum reduction determined in step S1303 for the selected typical pattern.

  FIG. 19 shows the processing in step S400 of the loudness measurement system according to the second modification of the first embodiment.

(S411)
The presentation sound control unit 220 determines a plurality of presentation sounds having different frequencies. Unlike S201, the presentation sound control unit 220 determines a presentation sound having a plurality of different sound pressures for each determined frequency.

(S412)
The presentation sound generation unit 120 generates the determined presentation sound. The electroacoustic transducer 130 converts the presentation sound generated by the presentation sound generation unit 120 into an acoustic signal and presents it to the user 100.

(S413)
The electroencephalogram measurement unit 150 measures an electroencephalogram corresponding to the potential difference between the plurality of electrodes 140 attached to the user 100.

(S414)
The electroencephalogram extraction unit 192 extracts an electroencephalogram having a predetermined time width from the electroencephalogram measured in S413 as a starting point. For example, the electroencephalogram extraction unit 192 extracts an electroencephalogram in a section of −100 ms or more and +400 ms or less using the presentation start point of the presentation sound.

(S1415)
The pattern expansion / contraction unit 191 expands / contracts the typical pattern using the function parameter that gives the maximum expansion and the parameter of the function that gives the minimum reduction output in S1306, and obtains a correlation coefficient with the electroencephalogram cut out in step S414. . The correlation coefficient between the extracted pattern and the extracted electroencephalogram is obtained by sequentially shifting the cutout position.

(S416)
The pattern matching unit 193 determines whether the maximum value among the correlation coefficients obtained in step S415 is equal to or greater than a reference value. This standard indicates the lower limit of the degree of similarity between the measured electroencephalogram and the typical pattern. For example, the reference is 0.5. If the maximum correlation coefficient is greater than or equal to the reference, the process proceeds to step S417. If the maximum correlation coefficient is less than the reference, it is determined that the brain wave is not adopted as data, and the process proceeds to step S418.

(S417)
The selection determination unit 194 specifies the expansion / contraction range and the time range of the typical pattern that maximizes the correlation coefficient obtained in S415. The selection determination unit 194 acquires the latency of the N1 component in the typical pattern when the correlation coefficient is maximized. The selection determination unit 194 determines the latency of the N1 component acquired using the typical pattern as the latency of the N1 component of the electroencephalogram measured in S413.

(S418)
The measurement completion determination unit 210 determines whether or not a predetermined number or more of N1 components of the electroencephalogram have been determined. The measurement completion determination unit 210 may hold a predetermined number for each frequency of the presentation sound. For example, the measurement completion determination unit 210 holds 5 as a predetermined number.

  When the measurement completion determination unit 210 determines that the N1 components of the brain waves equal to or greater than the predetermined number have been determined, the process proceeds to step S420. When the measurement completion determination unit 210 determines that N1 components of the brain waves equal to or greater than the predetermined number have not been determined, the process returns to step S411.

(S421)
The latency estimation unit 200 obtains the average latency of the N1 component for each frequency and sound pressure of the presented sound.

(S422)
The measurement completion determination unit 210 summarizes the latency of the N1 component for each sound pressure of the presentation sound for each frequency of the presentation sound, and generates a loudness increase curve for each frequency.

  The loudness measurement system configured as described above can be deformed according to the latency of each component of the electroencephalogram by matching the typical pattern by nonlinear expansion and contraction on the time axis. it can. As a result, the latency of the N1 component can be accurately measured with a small number of measurements, and the loudness increase curve can be measured in a short time.

(Embodiment 2)
FIG. 20 shows a configuration of the loudness measurement system 21 of the second embodiment. In the loudness measurement system 21 shown in FIG. 20, the pattern storage unit 170 of the loudness measurement system 10 of the first embodiment is replaced by the pattern storage units 470 a to 470 c, and the typical pattern selection unit 180 is replaced by the typical pattern selection unit 480. The configuration is the same except that the control unit 220 is replaced with a presentation sound control unit 420 and a switch 410 is added. The same components as those in the first embodiment are denoted by the same reference numerals, and description thereof is omitted as appropriate.

  The loudness measurement system 21 includes a presentation sound determination unit 110, a presentation sound generation unit 120, an electroacoustic conversion unit 130, an electroencephalogram measurement unit 150, a sample pattern generation unit 160, a typical pattern storage unit 470, and a typical pattern storage. 471, a typical pattern storage unit 472, a typical pattern selection unit 180, a switch 410, an electroencephalogram selection unit 190, a latency estimation unit 200, a measurement completion determination unit 210, and a presentation sound control unit 220. . In FIG. 1, the user 100 shown for convenience of explanation is not a component of the loudness measurement system 21.

  The loudness measurement system 21 is the same as the entire process shown in FIG. Processing different from the processing of the loudness measurement system according to the first embodiment will be described. FIG. 21 shows a flowchart of processing of the loudness measurement system 21 of the second embodiment. Description of operations similar to those in Embodiment 1 is omitted as appropriate.

(S100)
The loudness measurement system 10 starts auditory measurement after receiving an input to start measurement. For example, the presenting sound determination unit 110 receives an input for starting measurement.

(S200)
The presentation sound determination unit 110 determines a plurality of initial presentation sounds from the sounds stored in the presentation sound storage unit 111 with reference to the loudness correspondence relationship between the sound pressure value and the loudness value. The determined initial presentation sounds have different frequencies and sound pressures having the same loudness value. The presenting sound generation unit 120 generates information on the initial presenting sound including the order of presenting the presenting sound and the interval of presenting the presenting sound. The electroacoustic transducer 130 presents a presentation sound to the user 100 based on the generated information of the initial presentation sound. The electroencephalogram measurement unit 150 measures the electroencephalogram of the user 100 when the initial presentation sound is presented.

  In step S200, the operations from step S201 to step S205 shown in FIG. 7 are performed. The typical patterns stored in the typical pattern storage units 470 to 472 hold information on the presenting sound presentation interval. In S202, the presenting sound generation unit 120 receives the presenting sound presenting interval information from the storage units 470 to 472, and determines a presenting interval longer than those presenting intervals. Moreover, the presentation sound production | generation part 120 may hold | maintain the information made into a predetermined presentation interval or less. The presenting interval here may be an interval at which a plurality of presenting sounds having the same frequency are presented. For example, a presenting sound having a frequency of 1000 Hz (first sound), a presenting sound having a frequency of 4000 Hz (second sound), a presenting sound having a frequency of 2000 Hz (third sound), and a presenting sound having a frequency of 1000 Hz (fourth sound) are spaced at intervals of 500 ms. Let us consider the case where presentations are made in order. The time interval between the presentation sound having the first 1000 Hz (first sound) and the next presentation sound having the 1000 Hz (fourth sound) is 1500 ms (500 ms + 500 ms + 500 ms).

  In step S200 of the second embodiment, the sound is presented at a presentation interval or more corresponding to the largest presentation interval of the sounds stored in the typical pattern storage unit 471 from the typical pattern storage unit 470. For example, an electroencephalogram for a sound presented at a presentation interval of 3 seconds or more is measured.

  FIGS. 22A, 22B, and 22C show how the typical pattern storage unit 470 is deformed under the influence of the presentation interval of the typical pattern presentation sound stored in the typical pattern storage unit 472. FIG. It is known that the event-related potential changes when the user 100 can predict the sound presented. Therefore, a plurality of presentation sounds having different frequencies are presented randomly so that the user 100 cannot predict the presentation sound. At this time, the presentation interval of the presentation sound changes depending on the presentation order of the presentation sounds. It is known that the amplitudes of the N1 component and the P2 component change when the presentation interval of the presentation sound changes. 22 (a), 22 (b), and 22 (c), a waveform when the presentation interval of the presentation sound is 3 seconds or more, a waveform when the presentation interval is 2 seconds or more and shorter than 3 seconds, and the presentation interval is 1.5 seconds or more and shorter than 2 seconds. The waveform of each case is shown. For example, in the waveform shown in FIG. 22A, when the presentation interval is 3 seconds or more, the amplitude of the N1 component and the amplitude of the P2 component are approximately the same. However, the amplitude of the P2 component decreases as the waveform becomes shorter when it is shorter than 2 seconds and shorter than 3 seconds and when the presentation interval is shorter than 1.5 seconds and shorter than 2 seconds. Similarly to the waveform shown in FIG. 22A, the waveform shown in FIG. 22B also decreases in the amplitude of the P2 component as the presentation interval becomes shorter. In the waveform shown in FIG. 22C, the amplitude of the N1 component decreases as the presentation interval becomes shorter. For example, the typical pattern storage unit 470 stores the waveform shown in FIG. The typical pattern storage unit 471 stores the waveform shown in FIG. The typical pattern storage unit 472 stores the waveform shown in FIG.

(S2300)
The sample pattern generation unit 160 generates an electroencephalogram sample pattern by averaging the brain waves with respect to the initial presentation sound. The typical pattern selection unit 180 expands and contracts the typical patterns of a plurality of brain waves including the N1 component and the P2 component stored in the typical pattern storage unit 472 from the typical pattern storage unit 470 within a predetermined expansion / contraction range and time range. The typical pattern selection unit 180 compares the stretched typical pattern with the generated electroencephalogram sample pattern, and selects the typical pattern closest to the electroencephalogram sample pattern.

(S2400)
The presented sound control unit 220 determines a plurality of presented sounds having different frequencies and sound pressures. The electroacoustic transducer 130 presents the user 100 with a plurality of presenting sounds having different frequencies and sound pressures. The electroencephalogram measurement unit 150 measures the electroencephalogram of the user 100 when the presentation sound is presented.
The electroencephalogram selection unit 190 expands / contracts the typical pattern stored in one of the typical pattern storage unit 472 stored in the typical pattern storage unit 472 in the predetermined expansion / contraction range and time range. Compare with the measured electroencephalogram. The latency estimation unit 200 uses the latency of the N1 component acquired by the electroencephalogram selection unit 190 to estimate the latency of the N1 component for each frequency of the presentation sound and the sound pressure of the presentation sound. The measurement completion determination unit 210 refers to a predetermined relationship between the sound pressure of the presenting sound and the latency of the N1 component, and for each frequency of the presenting sound, the estimated N1 component latent for the sound pressure of the presenting sound. Find the loudness increase curve corresponding to the change of time.

(S500)
The measurement completion determination unit 210 outputs a loudness increase curve for each frequency and ends the measurement.

  FIG. 23 shows a detailed configuration of the typical pattern selection unit 480. FIG. 24 is a flowchart showing step S2300 of the processing of the loudness measurement system 21.

  The typical pattern matching unit 183 shown in FIG. 23 is the same as FIG. 4 except that the typical pattern matching unit 183 is replaced with a typical pattern matching unit 483. The typical pattern selection unit 480 includes a detection unit 181, a latency acquisition unit 182, a switch 410 that switches the typical pattern storage units 470, 471, and 472, and a pattern matching unit 483.

(S301)
The detection unit 181 determines the N1 component and the P2 component using the N1 component candidate and the P2 component candidate detected in step S205. When the detection unit 181 detects one N1 component candidate, the N1 component candidate is determined as the N1 component. When the detection unit 181 detects one P2 component candidate, the P2 component candidate is determined as the P2 component.

(S302)
The latency acquisition unit 182 acquires a latency difference that is a difference between the latency of the P2 component and the latency of the N1 component. The latency acquisition unit 182 acquires the time of the N1 component and the time of the P2 component. The time of the N1 component corresponds to the elapsed time from the time when the presentation sound is presented to the N1 component. The time of the P2 component corresponds to the elapsed time from the time when the presentation sound is presented to the P2 component.

(S2301)
The typical pattern matching 483 switches the switch 410 and connects any one of the typical pattern storage units 470 to 472. In the following description, it is assumed that the pattern matching unit 483 and the typical pattern storage unit 470 are connected.

(S2302)
The pattern matching unit 183 determines the expansion and contraction ranges of the N1 component and the P2 component included in the typical pattern stored in the typical pattern storage unit 470. The method for determining the expansion / contraction range is the same as S303 shown in FIG.

(S304)
The typical pattern matching unit 183 determines a time range in which the sample pattern of the electroencephalogram and the typical pattern are matched. This is the same as S304 shown in FIG.

(S2304)
The typical pattern matching unit 483 uses the determined expansion / contraction range and time range, and the phase relationship between a plurality of typical patterns having the largest presentation interval stored in the typical pattern storage unit 470 and the EEG sample pattern Find a number.

(S2305)
The pattern matching unit 483 confirms whether or not matching has been performed for all the typical pattern sequences included in the typical pattern storage units 470 to 472. If the pattern matching unit 483 obtains correlation coefficients between all the typical patterns included in the typical pattern storage units 470 to 472 and the electroencephalogram sample pattern, the process proceeds to step S2306. If the correlation coefficient between any of the typical patterns included in the N1-P2 typical pattern storage units 470 to 472 and the electroencephalogram sample pattern has not been obtained, the process returns to step S2301.

(S2306)
The typical pattern matching unit 183 refers to the correlation coefficient obtained in S305 and selects a typical pattern having the highest correlation coefficient.

  FIG. 25 is a flowchart showing details of step S2400 in the flowchart shown in FIG. Step S2400 includes a process for brain waves for each presentation sound presentation (S2410) and a process for collecting the results processed in step S410 for each type of presentation sound (S420).

  FIG. 25 is a flowchart showing details of the operation in step S2410, which is part of the operation of the present embodiment.

(S2411)
The typical pattern matching unit 483 included in the typical pattern selection unit 480 connects the switch 410 to one of the typical pattern storage units 470 to 472 that stores the typical pattern selected in step S2300. For example, the switch 410 is connected to the typical pattern storage unit 470.

(S2412)
The presentation sound control unit 420 determines one of the presentation sounds for which a predetermined loudness increase curve is measured and which has not been measured, and determines the output time of the presentation sound. . The presentation sound control unit 420 transmits the type of the presentation sound and the output time to the presentation sound generation unit 120 and the typical pattern selection unit 480.

(S412)
The presentation sound generation unit 120 generates the determined presentation sound. The electroacoustic transducer 130 presents the generated presentation sound to the user 100.

(S413)
The electroencephalogram measurement unit 150 measures an electroencephalogram corresponding to the potential difference between the plurality of electrodes 140 mounted on the scalp of the user 100. The measured electroencephalogram includes the time before and after the presentation time of the presentation sound.

(S414)
The electroencephalogram data selection unit 190 cuts out the electroencephalogram with a certain time width from the electroencephalogram recorded in step S413 with the presentation time of the presentation sound as a base point.

(S2414)
The typical pattern matching unit 483 included in the typical pattern selection unit 480 extracts a typical pattern corresponding to the typical pattern selected in step S2300 with reference to the presentation interval of the presentation sound, and transmits it to the electroencephalogram data selection unit 190. In step S2412, the typical pattern matching unit 483 generates the present sound from the time when the present sound having the same frequency is presented among the preceding present sounds based on the type of the present sound output from the present sound control unit 420 and the output time. The time interval until the presentation sound output time of the presentation sound is obtained. The obtained time interval is set as the presentation interval.

(S415)
The pattern expansion / contraction unit 191 included in the use electroencephalogram selection unit 190 expands / contracts the typical pattern extracted in step S2414 in the time direction, and the pattern matching unit 193 of the use electroencephalogram data selection unit 190 cuts out the contracted pattern in step S414. The correlation coefficient with the obtained EEG is obtained. The pattern expansion / contraction part 191 linearly expands / contracts the typical pattern, for example, 0.5 times or more and 1.5 times or less similarly to step S415 of the first embodiment.

  The pattern matching unit 193 extracts an electroencephalogram having the same time width as the typical pattern expanded and contracted by the pattern expansion / contraction unit 191 within the matching range of the electroencephalogram extracted in S414. As in the first embodiment, the matching range is determined in advance, and the range is, for example, from +70 ms to +450 ms, with the presentation sound being presented as 0. Similar to the first embodiment, the cut-out position is sequentially shifted to obtain the correlation coefficient between the expanded and contracted typical pattern and the cut-out electroencephalogram.

(S416)
In step S415, the pattern matching unit 193 determines whether or not the maximum value is greater than or equal to the reference value among the correlation coefficients calculated under each condition based on the combination of the expansion / contraction ratio of the typical pattern and the cutout section in the matching range. to decide. The standard defines the lower limit of the similarity between the measured electroencephalogram and the typical pattern. For example, 0.5. If the maximum correlation coefficient is greater than or equal to the reference, the process proceeds to step S417. If the maximum correlation coefficient is less than the reference, it is determined that the brain wave is not adopted as data, and the process proceeds to step S418.

(S417)
If the maximum correlation coefficient is greater than or equal to the reference in step S416, the selection determination unit 194 specifies the expansion / contraction rate of the typical pattern that maximizes the correlation coefficient and the electroencephalogram extraction range when calculating the correlation coefficient. The time position corresponding to the N1 peak of the typical pattern is specified on the time axis of the electroencephalogram from the expansion / contraction rate of the typical pattern and the cutout range. The position on the time axis corresponding to the peak of the N1 component of the typical pattern is defined as the latency of the N1 component of the electroencephalogram.

(S418)
The measurement completion determination unit 210 determines whether or not a predetermined number or more of brain waves have been adopted for each presentation sound for measuring a predetermined loudness increase curve. For example, the predetermined number is 5 or more. In step S418, when the number of adopted electroencephalograms satisfies the required number for all the presented sounds, the process proceeds to step S420. In step S418, if there is a presentation sound in which the number of brain waves adopted for the presentation sound does not satisfy the required number, the process returns to step S2412.

(S421)
If it is determined in step S418 that the measurement is completed, the measurement completion determination unit 210 averages the estimated value of the N1 latency of the electroencephalogram adopted for each type of the presented sound. The measurement completion determination unit 210 summarizes the average N1 latency calculated for each type of presented sound (including at least the sound pressure) for each frequency of the presented sound, and generates a loudness increase curve for each frequency. The loudness increase curve is obtained by the method shown in FIG. 9 described in the first embodiment.

  The loudness measurement system configured as described above performs initial measurement with a plurality of present sounds having different sound frequencies and the same loudness value, and a typical pattern of the auditory evoked potential according to individual characteristics. Select. A latency increase curve is generated by estimating the latency of the evoked potentials of individual brain waves that are not added and averaged by matching with the selected typical pattern. As a result, the latency of the evoked potential can be accurately measured with a small number of measurements, and the loudness increase curve can be measured in a short time.

  In the second embodiment, the initial measurement presentation interval in step S200 is, for example, 3 seconds or more. Furthermore, when selecting a typical pattern in step S2300, matching is performed between the typical pattern having the maximum presentation interval stored in the typical pattern storage units 470 to 472 and the sample pattern by the initial measurement.

  However, the initial measurement presentation interval in step S200 may be less than 3 seconds. In this case, that is, the initial presentation sound is presented at a time interval shorter than the maximum presentation interval among the presentation sound intervals included in the typical patterns stored in the N1-P2 typical pattern storage units 470 to 472. In step S2300, matching is performed using a typical pattern corresponding to the presentation interval of the initial measurement among the typical patterns stored in the typical pattern storage units 470 to 472. The typical pattern may not be linear expansion / contraction, but may be nonlinear expansion / contraction as in the second modification of the first embodiment.

  FIG. 29 shows the hardware configuration of the loudness measurement systems 10, 11, 12, and 21. The loudness measurement system includes a CPU 30, a memory 31, and an audio controller 32. The CPU 30, the memory 31, and the audio controller 32 are connected to each other via a bus 34 and can exchange data with each other.

  The CPU 30 executes a computer program 35 stored in the memory 31. The computer program 35 describes a processing procedure shown in a flowchart described later. The loudness measurement system performs processing such as determination of a presenting sound, measurement of an electroencephalogram, generation of a loudness increase curve, and the like according to the computer program 35. This process will be described in detail later.

  The audio controller 32 outputs a presenting sound via the electroacoustic transducer 130 in accordance with a command from the CPU 30.

  The loudness measurement system may be realized as hardware such as a DSP in which a computer program is incorporated in one semiconductor circuit. Such a DSP can realize all the functions of the CPU 30, the memory 31, and the audio controller 32 described above with a single integrated circuit.

  The computer program 35 described above can be recorded on a recording medium such as a CD-ROM and distributed as a product to the market, or can be transmitted through an electric communication line such as the Internet. A device (for example, a PC) having hardware shown in FIG. 29 can function as a loudness measurement system according to the present embodiment by reading the computer program 35.

  The loudness measurement system and the hearing aid adjustment system according to the present invention can be widely used when measuring auditory functions, and output adjustment of acoustic devices such as speakers and headphones, and devices that compensate auditory functions such as hearing aids This is useful when building a device that adjusts to the individual auditory condition.

10, 11, 12, 21 Loudness measurement system 100 User 110 Presented sound determination unit 120 Presented sound generation unit 130 Electroacoustic transducer 140 Electrode 150 EEG measurement unit 160 Sample pattern generation unit 170, 470a, 470b, 470c Typical pattern storage unit 180 480 Typical pattern selection unit 181 Detection unit 182 Latency acquisition unit 183, 483 Typical pattern matching unit 190 EEG selection unit 191 Pattern expansion / contraction unit 192 EEG extraction unit 193 Pattern matching unit 194 Selection determination unit 200 Latency estimation unit 210 Measurement completion determination Unit 220, 420 present sound control unit 320 HTL input unit 330 characteristic setting unit 340 hearing aid 410 switch

Claims (13)

Presenting the user with a plurality of initial presentation sounds having different pressures and sound pressures having the same loudness, which is a sensory sound level received by humans, and
Measuring a user's brain wave when presenting the initial presentation sound;
While expanding and contracting the typical pattern of the electroencephalogram including the event-related potential stored in the typical pattern storage unit, the event-related potential after a predetermined time from the time of the plurality of initial presentation sounds of the typical pattern and the measured electroencephalogram And a typical pattern selection step of selecting a typical pattern similar to the event-related potential;
Presenting to the user a plurality of presentation sounds having different frequencies and different sound pressures;
Measuring the user's brain wave when presenting the presenting sound;
EEG selection that expands and contracts the selected typical pattern and acquires the N1 component included in the typical pattern expanded and contracted so as to be most similar to the event-related potential after a predetermined time from the time of the plurality of presentation sounds among the brain waves Steps,
With reference to a predetermined relationship between the latency of the N1 component and the loudness, the loudness value corresponding to the acquired latency of the N1 component is used as the sound pressure of the presented sound when the electroencephalogram containing the N1 component is measured. Obtaining a loudness increase curve as a loudness measuring method. The typical pattern of an electroencephalogram including an event-related potential is information that associates an elapsed time after the sound is presented with the potential,
The typical pattern selection step expands and contracts in the time direction of the typical pattern, and compares the typical pattern and the event-related potential after a predetermined time from the time of the plurality of initial presentation sounds among the measured electroencephalograms.
The loudness measuring method according to claim 1. The event-related potential includes an N1 component that is a negative component after a range of 50 ms to 150 ms from the time when sound is presented, and a positive component that is a positive component after a range of 150 ms to 300 ms from the time when sound is presented. Including ingredients,
The loudness measuring method according to claim 1. In the typical pattern selection step, the difference between the latency of the N1 component and the latency of the P2 component included in the event-related potential after a predetermined time from the time of the plurality of initial presentation sounds is expanded or contracted to a predetermined magnification. Compare the typical pattern and the event-related potential after a predetermined time from the time of the plurality of initial presentation sounds of the measured electroencephalogram,
The loudness measuring method according to claim 3. In the typical pattern selection step, the plurality of initial presentation sounds of the typical pattern and the measured electroencephalograms are expanded and contracted so that the expansion / contraction rate increases every time from the time of the plurality of initial presentation sounds. Compare the event-related potential after a predetermined time from the time of
The loudness measuring method according to claim 3. The typical pattern selection step compares the event-related potential and the typical pattern included in a time range determined using the time of the N1 component and the time of the P2 component included in the event-related potential.
The loudness measuring method according to claim 3. The step of obtaining a loudness increase curve includes, for each frequency of the presentation sound, a loudness value corresponding to the latency of the N1 component acquired from a plurality of presentation sounds having a plurality of different sound pressures, and the N1 component. The loudness increase curve is obtained as the sound pressure of the presented sound when measuring the electroencephalogram.
The loudness measuring method according to claim 1. The step of presenting the plurality of initial presentation sounds to the user presents the initial presentation sounds at intervals greater than or equal to a presentation interval of a typical pattern of an electroencephalogram including an event-related potential stored in the typical pattern storage unit. ,
The loudness measuring method according to claim 1. A loudness measuring method according to claim 1;
The reference sound corresponding to the value of the loudness corresponding to the latency of the N1 component obtained in the step of obtaining the loudness increase curve with reference to the relationship between the latency of the N1 component, the loudness and the sound pressure of the reference sound. A setting step of setting a difference between the sound pressure of the determined reference sound and the sound pressure of the presentation sound when the N1 component is acquired;
Adjusting the hearing aid based on the difference set in the setting step. Multiple initial presentation sounds having different sound pressures that have the same loudness, which is the loudness of a sensory sound received by humans, and different sound pressures having different sound pressures A presentation unit that presents a plurality of presentation sounds to a user, an electroencephalogram measurement unit that measures the brain waves of the user,
A typical pattern storage unit storing a plurality of typical electroencephalogram patterns including event-related potentials;
While expanding and contracting the typical pattern, the typical pattern similar to the event-related potential is selected by comparing with the event-related potential after a predetermined time from the time of the plurality of initial presentation sounds among the typical pattern and the electroencephalogram A typical pattern selector,
An electroencephalogram that expands and contracts the selected typical pattern and acquires the latency of the N1 component contained in the expanded and contracted typical pattern that is most similar to the event-related potential after a predetermined time from the time of the plurality of presentation sounds in the electroencephalogram A selection section;
Referring to the relationship between the predetermined N1 component latency, the sound pressure of the presenting sound, and the loudness, and measuring the electroencephalogram including the N1 component, the loudness value corresponding to the acquired N1 component latency is measured. A measurement completion determination unit for obtaining a loudness increase curve as the sound pressure of the presenting sound,
Equipped with a loudness measurement system. A loudness measurement system according to claim 10;
The reference sound corresponding to the value of the loudness corresponding to the latency of the N1 component obtained in the step of obtaining the loudness increase curve with reference to the relationship between the latency of the N1 component, the loudness and the sound pressure of the reference sound. A hearing aid adjustment comprising: a characteristic setting unit that determines a difference between a sound pressure of the determined reference sound and a sound pressure of a presentation sound when the N1 component is acquired as a characteristic of adjustment of the hearing aid system. A computer program executed by a computer,
The computer program is for the computer.
Presenting a user with a plurality of initial presentation sounds having different pressures and sound pressures having the same loudness, which is a sensory sound level received by humans;
Measuring a user's brain wave when presenting the initial presentation sound;
While expanding and contracting the typical pattern of the electroencephalogram including the event-related potential stored in the typical pattern storage unit, the event-related potential after a predetermined time from the time of the plurality of initial presentation sounds of the typical pattern and the measured electroencephalogram And a typical pattern selection step of selecting a typical pattern similar to the event-related potential;
Presenting to the user a plurality of present sounds having different frequencies and different sound pressures;
Measuring the user's brain wave when presenting the presenting sound;
EEG selection that expands and contracts the selected typical pattern and acquires the N1 component included in the typical pattern expanded and contracted so as to be most similar to the event-related potential after a predetermined time from the time of the plurality of presentation sounds among the brain waves Steps,
With reference to a predetermined relationship between the latency of the N1 component and the loudness, the loudness value corresponding to the acquired latency of the N1 component is used as the sound pressure of the presented sound when the electroencephalogram containing the N1 component is measured. And a step of obtaining a loudness increase curve as a computer program. A computer program executed by a computer,
The computer program is for the computer.
Presenting a user with a plurality of initial presentation sounds having different pressures and sound pressures having the same loudness, which is a sensory sound level received by humans;
Measuring a user's brain wave when presenting the initial presentation sound;
While expanding and contracting the typical pattern of the electroencephalogram including the event-related potential stored in the typical pattern storage unit, the event-related potential after a predetermined time from the time of the plurality of initial presentation sounds of the typical pattern and the measured electroencephalogram And a typical pattern selection step of selecting a typical pattern similar to the event-related potential;
Presenting to the user a plurality of present sounds having different frequencies and different sound pressures;
Measuring the user's brain wave when presenting the presenting sound;
EEG selection that expands and contracts the selected typical pattern and acquires the N1 component included in the typical pattern expanded and contracted so as to be most similar to the event-related potential after a predetermined time from the time of the plurality of presentation sounds among the brain waves Steps,
With reference to a predetermined relationship between the latency of the N1 component and the loudness, the loudness value corresponding to the acquired latency of the N1 component is used as the sound pressure of the presented sound when the electroencephalogram containing the N1 component is measured. Obtaining a loudness increase curve as
The reference sound corresponding to the value of the loudness corresponding to the latency of the N1 component obtained in the step of obtaining the loudness increase curve with reference to the relationship between the latency of the N1 component, the loudness and the sound pressure of the reference sound. A setting step of setting a difference between the sound pressure of the determined reference sound and the sound pressure of the presentation sound when the N1 component is acquired;
A computer program for executing a step of adjusting a hearing aid based on the difference set in the setting step.

Images