Listening device, listening method, and program

Abstract

To provide a listening device capable of properly adjusting an output voice depending on the phoneme included in an input voice.SOLUTION: A listening device 10 involved in the present disclosure comprises a reference frequency band acquisition part 152, an amplification factor acquisition part 154, a microphone (a voice input part) 11, an analysis part 12, and an adjusting part 155: a reference frequency band acquisition part acquires the reference frequency band every phoneme, which includes at least one of a first frequency band to improve the recognition accuracy of a single word by increasing a sound pressure and a second frequency band to improve the recognition accuracy by lowering the sound pressure; an amplification factor acquisition part acquires the sound pressure amplification factor corresponding to the reference frequency band determined by a user; a microphone (a voice input part) inputs the audio around the user; an analysis part analyzes the voice and to extract the phoneme included in the voice; and an adjusting part adjusts the output voice based on the reference frequency band, the amplification factor, and the extracted phoneme.SELECTED DRAWING: Figure 4

Description

[Technical field]

[0001] The present disclosure relates to a hearing device, a hearing method, and a program. [Background technology]

[0002] There is known a technique for adjusting the volume level of a sound output from a speaker in accordance with a user's hearing level. Such a technique is used in, for example, hearing aids.

[0003] As a related technique, Patent Document 1 discloses a method for adjusting the audio output of a device according to a listener's desired level. In the method disclosed in Patent Document 1, a user sets the level of amplification for the center of each frequency band and test frequencies for a number of audio frequency bands. The control module of the device generates sounds at the test frequencies, and the user adjusts the amplification of each sound until the user is satisfied with hearing the sound. Other frequencies in the band are adjusted by the same amount as the test sound. In this way, the method disclosed in Patent Document 1 increases the amplification of frequencies that are difficult for the user to hear and limits the amplification of frequencies that are heard strongly. [Prior art documents] [Patent documents]

[0004] [Patent Document 1] Special Publication No. 2007-508751 Summary of the Invention [Problem to be solved by the invention]

[0005] In a conversation, whether a listener can accurately recognize words contained in a speaker's voice depends not only on the listener's hearing level but also on the phonemes contained in the speaker's voice. A phoneme is the smallest unit that separates sounds in the language used by the speaker. Therefore, even if the sound pressure of a frequency band that is difficult for a user to hear is increased using a hearing aid or the like, the user may not be able to accurately recognize words due to the influence of the phonemes. The technology disclosed in Patent Document 1 does not fully consider such problems.

[0006] In view of the above-mentioned problems, an object of the present disclosure is to provide a hearing device, a hearing method, and a program capable of appropriately adjusting an output sound according to phonemes contained in an input sound. [Means for solving the problem]

[0007] The hearing device according to the present disclosure comprises: a reference frequency band acquisition unit that acquires, for each phoneme, a reference frequency band including at least one of a first frequency band for improving word recognition accuracy by increasing sound pressure and a second frequency band for improving the recognition accuracy by decreasing sound pressure; an amplification factor acquisition unit that acquires an amplification factor of a sound pressure corresponding to the reference frequency band determined according to a user; A voice input unit for inputting voices around the user; an analysis unit that analyzes the speech and extracts phonemes contained in the speech; and an adjustment unit that adjusts an output voice based on the reference frequency band, the amplification factor, and the extracted phonemes.

[0008] The listening method according to the present disclosure comprises: The computer a reference frequency band acquisition step of acquiring, for each phoneme, a reference frequency band including at least one of a first frequency band for improving word recognition accuracy by increasing sound pressure and a second frequency band for improving the recognition accuracy by decreasing sound pressure; an amplification factor acquisition step of acquiring an amplification factor of a sound pressure corresponding to the reference frequency band determined according to a user; a voice input step of inputting voice around the user; an analysis step of analyzing the speech and extracting phonemes contained in the speech; and an adjustment step of adjusting an output voice based on the reference frequency band, the amplification factor, and the extracted phonemes.

[0009] The program according to the present disclosure is a reference frequency band acquisition step of acquiring, for each phoneme, a reference frequency band including at least one of a first frequency band for improving word recognition accuracy by increasing sound pressure and a second frequency band for improving the recognition accuracy by decreasing sound pressure; an amplification factor acquisition step of acquiring an amplification factor of a sound pressure corresponding to the reference frequency band determined according to a user; a voice input step of inputting voice around the user; an analysis step of analyzing the speech and extracting phonemes contained in the speech; and an adjustment step of adjusting an output voice based on the reference frequency band, the amplification factor, and the extracted phonemes. Effect of the Invention

[0010] The hearing device, hearing method, and program according to the present disclosure can appropriately adjust the output sound according to the phonemes contained in the input sound. [Brief description of the drawings]

[0011] [Figure 1] 1 is a block diagram showing an overall configuration of a system according to an embodiment. [Diagram 2] FIG. 2 is a block diagram showing a functional configuration of a server according to an embodiment. [Diagram 3] 1 is a block diagram showing a functional configuration of a hearing device according to an embodiment. [Figure 4]FIG. 2 is a block diagram showing a detailed configuration of a processing unit according to the embodiment. [Diagram 5] 11 is a flowchart showing a flow of a reference frequency band determination process performed by a server according to an embodiment. [Figure 6] 4 is a flowchart showing the flow of a setting process performed by the hearing device according to the embodiment. [Figure 7] 4 is a flowchart showing the flow of a sound adjustment process performed by the hearing device according to the embodiment. [Figure 8] FIG. 13 is a diagram showing frequency characteristics before adjustment according to the embodiment. [Figure 9] FIG. 13 is a diagram showing frequency characteristics after adjustment according to the embodiment. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0012] Hereinafter, embodiments of the present disclosure will be described in detail with reference to the drawings. In each drawing, the same or corresponding elements are denoted by the same reference numerals. For clarity of explanation, duplicate explanations will be omitted as necessary.

[0013] (Configuration and Overview of System 100) The configuration and overview of a system 100 according to an embodiment will be described with reference to Fig. 1. Fig. 1 is a block diagram showing the overall configuration of the system 100. The system 100 includes a server 20 and a hearing device 10. Although one server 20 and one hearing device 10 are shown in Fig. 1, the number of servers 20 and the number of hearing devices 10 are not limited to those shown in the figure. The number of servers 20 may be any number equal to or greater than one, and the number of hearing devices 10 may be any number equal to or greater than one.

[0014] The server 20 and the hearing device 10 each include a communication unit not shown in Fig. 1, and are connected via a network N. The network N is a wireless or wired communication line. As the network N, for example, an Internet line, a mobile phone line, a wireless LAN (Local Area Network), a wired LAN, etc. may be used, and the type of communication is not limited.

[0015] The server 20 is, for example, a PC (Personal Computer), etc. The server 20 transmits to the audio hearing device 10 various types of information used in the processing in the audio hearing device 10.

[0016] The hearing device 10 is a hearing device used by a user of the system 100. The user uses the hearing device 10, for example, to improve the ease of hearing surrounding sounds. The user may be, for example, a person with hearing loss whose hearing level is below a predetermined value. However, the user may also be a person with normal hearing whose hearing level is equal to or higher than a predetermined value.

[0017] The hearing device 10 is, for example, a hearing aid. The hearing device 10 may be, for example, an over-the-head headphone having a headband, or a canal-type or inner-ear type earphone. The hearing device 10 may be a left-right separated type (completely wireless type) earphone in which the left and right output units are independent. The hearing device 10 may be a left-right integrated type earphone such as a neckband type to which the left and right output units are connected. The hearing device 10 may be a portable speaker that can be placed around the user. The above is an example, and the hearing device 10 is not limited to the above example. It is to be noted that the hearing device 10 is not limited to being worn on both the left and right ears. The hearing device 10 may be configured to be worn on either the left or right ear.

[0018] Here, a hearing aid will be used as an example of the hearing device 10. A user wears the hearing device 10 on each of the left and right ears. Note that the hearing devices 10 worn on the left and right ears may have the same configuration or different configurations.

[0019] The system 100 is a system capable of improving the hearing of a user by performing a predetermined process by the hearing device 10. The hearing device 10 performs the predetermined process using information received from the server 20 and information acquired by the hearing device 10.

[0020] Specifically, the hearing device 10 acquires a voice around the user, extracts phonemes contained in the acquired voice, adjusts the voice according to the extracted phonemes, and outputs the adjusted voice. The hearing device 10 adjusts the output voice, for example, by increasing the sound pressure of a predetermined frequency band or decreasing the sound pressure of a predetermined frequency band. The predetermined frequency band is, for example, a reference frequency band obtained by investigating a plurality of subjects. The reference frequency band includes at least one of a first frequency band for improving word recognition accuracy by increasing sound pressure and a second frequency band for improving recognition accuracy by decreasing sound pressure.

[0021] The "word recognition accuracy" indicates the accuracy of the user's recognition of words contained in the voice. Note that the recognition accuracy is not limited to one word, and may be the recognition accuracy of a phrase including multiple words. Note that the subject may be, for example, a user of the system 100.

[0022] The first frequency band is a frequency band that increases the sound pressure to improve the word recognition accuracy of multiple subjects. In other words, the hearing device 10 adjusts the output sound to increase the sound in the first frequency band, thereby promoting word recognition in the user. Therefore, hereinafter, the first frequency band may be referred to as the "promotion frequency band."

[0023] On the other hand, the second frequency band is a frequency band that improves the word recognition accuracy of multiple subjects by lowering the sound pressure. That is, the hearing device 10 adjusts the output sound to reduce the sound of the second frequency band, thereby promoting word recognition in the user. In other words, when the hearing device 10 adjusts the output sound to increase the sound of the second frequency band, it suppresses the user's word recognition. Therefore, hereinafter, the second frequency band may be referred to as a "suppression frequency band". In addition, hereinafter, an example will be described in which the reference frequency band includes both the promotion frequency band and the suppression frequency band. In addition, in general, the promotion frequency band and the suppression frequency band do not overlap.

[0024] The reference frequency band is, for example, predetermined in the server 20. The determination method will be described later. The server 20 transmits information on the predetermined reference frequency band to the hearing device 10. The hearing device 10 receives the information on the reference frequency band from the server 20 and adjusts the output sound using the reference frequency band. For example, the hearing device 10 adjusts the input sound by increasing the sound pressure of the promotion frequency band of the input sound and decreasing the sound pressure of the suppression frequency band, and outputs the adjusted sound. This allows the hearing device 10 to improve the user's word recognition accuracy.

[0025] In addition, the hearing device 10 may further adjust the sound according to the user. For example, the hearing device 10 acquires an amplification factor of the sound pressure corresponding to the reference frequency band determined according to the user, and adjusts the output sound based on the reference frequency band, the amplification factor, and the extracted phonemes. The amplification factor is different from the reference frequency band determined based on the data of multiple subjects, and is determined for each user according to the individual state of the user. This allows the hearing device 10 to adjust the output sound based on the amplification factor according to the user. In this way, the system 100 can output to the user a sound that is appropriately adjusted according to the phonemes contained in the sound input to the hearing device 10.

[0026] (Server 20 configuration) The configuration of the server 20 will be described in detail with reference to Fig. 2. Fig. 2 is a block diagram showing the functional configuration of the server 20. The server 20 may be a dedicated device for performing the processing according to this embodiment, or may be a general-purpose device.

[0027] The server 20 includes a recognition rate calculation unit 21, a reference frequency band determination unit 22, and a communication unit 24. The server 20 includes a processor, a memory, and a storage device, which are not shown in the figure. The storage device stores a computer program in which the processing according to this embodiment is implemented. The processor can load the computer program from the storage device into the memory and execute the computer program. In this way, the processor realizes the functions of the recognition rate calculation unit 21 and the reference frequency band determination unit 22. Alternatively, the recognition rate calculation unit 21 and the reference frequency band determination unit 22 may each be realized by dedicated hardware.

[0028] In addition, some or all of the components of each device may be realized by general-purpose or dedicated circuits, processors, etc., or a combination of these. These may be configured by a single chip, or by multiple chips connected via a bus. Some or all of the components of each device may be realized by a combination of the above-mentioned circuits, etc., and programs. In addition, a CPU (Central Processing Unit), a GPU (Graphics Processing Unit), an FPGA (field-programmable gate array), a quantum processor (quantum computer control chip), etc. can be used as the processor.

[0029] Furthermore, when some or all of the components of the server 20 are realized by a plurality of information processing devices, circuits, etc., the plurality of information processing devices, circuits, etc. may be centrally arranged or distributed. For example, the information processing devices, circuits, etc. may be realized as a client-server system, a cloud computing system, etc., in a form in which each is connected via a communication network. Furthermore, the functions of the server 20 may be provided in a SaaS (Software as a Service) format. The above-mentioned method of realizing each component of the server 20 is also the same for the hearing device 10.

[0030] The recognition rate calculation unit 21 acquires information on a plurality of subjects and calculates a standard recognition rate for each phoneme. For example, the phonemes include consonants and vowels such as d, b, in, o, l, a, r, p, h, g, ch, sh, t, f, th, and s, which are concentrated in the frequency band from 500 Hz to 4000 Hz.

[0031] The recognition rate calculation unit 21 has a plurality of subjects (for example, 100 or more people) listen to unadjusted words whose sound pressure has not been adjusted. The recognition rate calculation unit 21 acquires a recognition result indicating how the unadjusted words sounded from each of the plurality of subjects. The recognition rate calculation unit 21 may acquire the recognition result via the hearing device 10, or may acquire the recognition result by another method. The recognition rate calculation unit 21 judges whether the recognition result is correct or incorrect. The recognition rate calculation unit 21 calculates the recognition rate of the words based on the number of subjects and the number of subjects who answered correctly. The subjects who answered correctly are subjects who correctly recognized the words. In this way, the recognition rate calculation unit 21 calculates the recognition rate of the unadjusted words for the plurality of subjects. The recognition rate calculation unit 21 also calculates the recognition rate of the words whose sound pressure has been adjusted in the same manner as described above. A specific example of the calculation of the recognition rate will be described later.

[0032] The reference frequency band determination unit 22 determines, for each phoneme, a reference frequency band including at least one of a promotion frequency band (first frequency band) for improving word recognition accuracy by increasing sound pressure and a suppression frequency band (second frequency band) for improving recognition accuracy by decreasing sound pressure. Here, the reference frequency band determination unit 22 determines both the promotion frequency band and the suppression frequency band as the reference frequency band.

[0033] For example, suppose a user hears the word "shuppatsu (departure)" in a conversation. In an environment with strong high-frequency noise, the background noise may exceed the sound pressure (level) of the "shuuu" sound, causing the user to mistakenly hear "shuuu" as "batu." In such a case, by lowering the sound pressure of the frequency band adjacent to the center frequency of the "shuuu" sound, the user can more easily hear the "shuuu" sound. Therefore, the frequency band adjacent to the center frequency of the "shuuu" sound is a suppression frequency band.

[0034] In addition, by increasing the sound pressure in the center of the frequency band of the whoosh sound, the user can more easily hear the word. In this case, the center of the frequency band of the whoosh sound is the facilitating frequency band. Therefore, by increasing the sound pressure in the center to a level that the user can at least hear, the user's word recognition accuracy is improved.

[0035] The reference frequency band determination unit 22 determines the reference frequency band by performing a reference frequency band determination process described later. The reference frequency band determination unit 22 determines whether the word recognition accuracy of multiple subjects is improved by changing the sound pressure in each of multiple frequency bands divided into a predetermined range. For example, the reference frequency band determination unit 22 can make this determination using a recognition rate that indicates the proportion of subjects who correctly recognize words among the multiple subjects.

[0036] The reference frequency band determination unit 22 identifies a frequency band for improving the recognition accuracy according to the judgment result, and determines the identified frequency band as the reference frequency band. The reference frequency band determination unit 22 transmits the determined reference frequency band to the hearing device 10. In the hearing device 10, the reference frequency band is recorded in the second memory M2.

[0037] In this way, in the system 100, the server 20 performs a primary judgment by determining whether an unspecified number of subjects can hear words correctly. Also, as described later, in the hearing device 10, the sound pressure of a specific frequency band is adjusted according to the user, and the tuning point at which hearing is improved is optimized. In this way, the system 100 can make the interrelationship between the promotion frequency band and the inhibition frequency band into an algorithm and a module.

[0038] The communication unit 24 communicates with the hearing device 10. The communication unit 24 may communicate with a plurality of hearing devices 10. The communication unit 24 may also communicate with devices other than the hearing device 10. The communication unit 24 may be configured with a communication interface for performing wired or wireless communication. The specific flow of processing performed by the server 20 will be described later.

[0039] (Configuration of hearing device 10) The configuration of the hearing device 10 will be described in detail with reference to Fig. 3. Fig. 3 is a block diagram showing the functional configuration of the hearing device 10. The hearing device 10 comprises a microphone 11, an analysis unit 12, an AI (Artificial Intelligence) 13, a communication unit 14, a processing unit 15, a speaker 16, a first memory M1, a second memory M2, a third memory M3, and a fourth memory M4.

[0040] In the figure, one-way arrows shown by solid or dashed lines simply indicate the flow of information (data, signals, etc.) and do not exclude the two-way flow of information. The same applies to Figure 4 described later.

[0041] The hearing device 10 includes a processor, a memory, and a storage device, which are not shown in the figure. The memory and the storage device may each be realized by any one of the first memory M1 to the fourth memory M4 shown in the figure. A computer program in which the processing according to this embodiment is implemented is stored in the storage device. The processor can load the computer program from the storage device into the memory and execute the computer program. In this way, the processor realizes the functions of the analysis unit 12, the AI ​​13, and the processing unit 15. Alternatively, the analysis unit 12, the AI ​​13, and the processing unit 15 may each be realized by dedicated hardware.

[0042] In addition, some or all of the components of the hearing device 10 may be realized by general-purpose or dedicated circuits, processors, etc., or a combination of these. Specific examples of how to realize the components of the hearing device 10 are the same as those of the server 20 described above, so detailed explanations will be omitted here.

[0043] The microphone 11 is an example of a voice input unit that inputs voices around the user. The voices around the user are, for example, the voice of the user, the voice of the person the user is talking to, or environmental sounds around the user. The microphone 11 converts the input voice into voice data and outputs it to the analysis unit 12. The microphone 11 may be a microphone that is provided in advance in the hearing device 10, or may be an external microphone connected to an input interface (not shown). In addition, the microphone 11 is not limited to the microphone 11, and other voice input devices may be used as the voice input unit.

[0044] The analysis unit 12 analyzes the voice input from the microphone 11 and extracts phonemes contained in the voice. The analysis unit 12 may extract phonemes using a known analysis technique. The analysis unit 12 may also analyze the voice taking into account environmental sounds contained in the input voice and extract phonemes. By receiving a notification regarding environmental sounds from the AI ​​13, the analysis unit 12 can analyze the voice taking into account the inclusion of environmental sounds in the input voice. The analysis unit 12 outputs the analysis result to the processing unit 15.

[0045] The analysis unit 12 may also receive a notification regarding the language from the AI ​​13 and switch the language from which phonemes are extracted. This allows the analysis unit 12 to perform analysis according to the language used in the user's conversation. In this case, the analysis unit 12 notifies the processing unit 15 of information regarding the switched language.

[0046] The AI ​​13 analyzes the voice input from the microphone 11 from a different perspective than the analysis unit 12. For example, the AI ​​13 analyzes the input voice according to the user's reaction in the conversation and the user's surrounding environment. The AI ​​13 outputs the analysis result to the analysis unit 12. This allows the analysis unit 12 to further analyze the voice by taking into account the analysis result in the AI ​​13.

[0047] For example, the AI ​​13 analyzes the input voice using a trained model generated in advance by a training process. The training algorithm is not particularly limited. For example, a known method such as a convolutional neural network (CNN) using a deep learning technique using a neural network may be used.

[0048] For example, the AI ​​13 determines the user's word recognition status based on the content of the user's speech. Specifically, the AI ​​13 first acquires the user's voice from the microphone 11 while he or she is talking to people around him or her. The AI ​​13 analyzes the acquired user's voice and extracts from the user's voice words that indicate repetition of the same words or asking the speaker to repeat the same words. For example, the AI ​​13 extracts words such as "Eh?" and "What is it?"

[0049] When repetition of the same word or asking the speaker to repeat the word is extracted, AI13 determines that the user did not hear the word. In this case, AI13 records data of the word (phoneme) that the user did not hear in the fourth memory M4. In addition, AI13 sends a notification to the analysis unit 12 to update the amplification factor (described later) according to the determination result. As a result, the analysis unit 12 sends the notification to the processing unit 15. In addition, the processing unit 15 can update the amplification factor based on the notification.

[0050] The AI13 may make the above determination using a trained model that has been trained in advance. For example, the AI13 analyzes the voice using a trained model that has been trained to receive feature data of a voice signal indicating the user's speech as an input and to receive the user's word recognition status as an output. The AI13 inputs the acquired voice into the trained model and receives the user's word recognition status as an output. In this way, the AI13 can determine the user's word recognition status using the trained model. The AI13 may retrain the trained model. This allows the AI13 to more accurately determine the user's recognition status.

[0051] Furthermore, the AI ​​13 may analyze the language used by the speaker input through the microphone 11, and notify the analysis result to the analysis unit 12. This causes the analysis unit 12 to switch the language used for analysis.

[0052] Furthermore, AI13 may perform an analysis of environmental sounds contained in the acquired voice. For example, AI13 extracts environmental sounds from the acquired voice. AI13 notifies the analysis unit 12 of information on the extracted environmental sounds. As a result, the analysis unit 12 performs an analysis that takes into account the environmental sounds. Note that, although the configuration of AI13, which is an artificial intelligence, has been described here, the hearing device 10 may perform the above-mentioned processing of AI13 using a function other than artificial intelligence.

[0053] The communication unit 14 communicates with the server 20. The communication unit 14 may communicate with a device other than the server 20. The communication unit 14 may be configured with a communication interface for performing wired or wireless communication.

[0054] The processing unit 15 refers to the first memory M1 to the fourth memory M4 and acquires information stored therein. The processing unit 15 performs a voice adjustment process to adjust the output voice based on the acquired information. The processing unit 15 also receives a notification regarding the language from the analysis unit 12 and switches the reference frequency band used in the voice adjustment process to a reference frequency band corresponding to the language used by the speaker.

[0055] 4 is a block diagram showing a detailed configuration of the processing unit 15. The processing unit 15 includes an initial setting unit 151, a reference frequency band acquisition unit 152, an amplification factor determination unit 153, an amplification factor acquisition unit 154, and an adjustment unit 155.

[0056] The initial setting unit 151 performs initial settings on the hearing device 10. For example, the initial setting unit 151 identifies a frequency band in which the recognition accuracy of the user's words is insufficient, regardless of the phonemes contained in the input voice. This frequency band may be referred to as a "specific frequency band" below. The initial setting unit 151 may identify the specific frequency band using a known method. The initial setting unit 151 records information on the specific frequency band in the first memory M1.

[0057] In addition, insufficient word recognition accuracy in a specific frequency band can be addressed to some extent by, for example, increasing the sound pressure of the specific frequency band in a known hearing aid. In a known hearing aid, the sound pressure of a specific frequency band is increased without considering the influence of phonemes to compensate for the user's hearing in the specific frequency band. Here, an example will be described in which the hearing device 10 also has a hearing aid function that such a known hearing aid has, but this is not limited to this. The hearing device 10 does not need to have such a known hearing aid function. In other words, the hearing device 10 does not need to have an initial setting unit 151 that specifies a specific frequency band.

[0058] The reference frequency band acquisition unit 152 acquires, for each phoneme, a reference frequency band including at least one of a promotion frequency band (first frequency band) for improving the word recognition accuracy by increasing the sound pressure and a suppression frequency band (second frequency band) for improving the recognition accuracy by decreasing the sound pressure. Here, the reference frequency band acquisition unit 152 acquires both the promotion frequency band and the suppression frequency band as the reference frequency band.

[0059] As described above, the reference frequency band is determined by the reference frequency band determination unit 22 of the server 20. The server 20 also transmits the determined reference frequency band to the hearing device 10 and records the reference frequency band in the second memory M2. The reference frequency band acquisition unit 152 acquires the reference frequency band by referring to the second memory M2.

[0060] The amplification factor determination unit 153 determines the amplification factor of the sound pressure corresponding to the reference frequency band according to the user. The amplification factor determination unit 153 determines the amplification factor of the promotion frequency band to be greater than 1 and the amplification factor of the suppression frequency band to be less than 1. The amplification factor determination unit 153 records the determined amplification factor in the third memory M3.

[0061] Furthermore, the amplification factor determination unit 153 may function as an update unit that updates the amplification factor according to the user's word recognition status. For example, the amplification factor determination unit 153 acquires an analysis result in the AI ​​13, and updates the amplification factor based on the acquired analysis result. This allows the amplification factor determination unit 153 to dynamically change the amplification factor according to the user's word recognition status. The amplification factor determination unit 153 may update the amplification factor in real time. This allows the user to always have a conversation with a voice that is easy to hear.

[0062] The amplification factor acquisition unit 154 refers to the third memory M3 and acquires the amplification factor of the sound pressure corresponding to the reference frequency band, which is determined according to the user. Note that, here, an example will be described in which the amplification factor determination unit 153 determines the amplification factor, and the determined amplification factor is acquired by the amplification factor acquisition unit 154, but this is not limiting. For example, a device other than the hearing device 10 may determine the amplification factor, and the determined amplification factor may be acquired by the amplification factor acquisition unit 154. In this case, the processing unit 15 may not include the amplification factor determination unit 153.

[0063] The adjustment unit 155 adjusts the output sound. Here, the "output sound" refers to the sound to be output from the speaker 16. The adjustment unit 155 causes the speaker 16 to output the adjusted output sound.

[0064] For example, the adjustment unit 155 adjusts the output sound so as to increase the sound pressure of a specific frequency band. As a result, the adjustment unit 155 adjusts the output sound for a frequency band in which the user has difficulty hearing words, similar to the hearing aid function of a general hearing aid. This adjustment does not take into account the phonemes contained in the sound, but can be said to be adjustment according to the hearing level of the user.

[0065] The adjustment unit 155 also adjusts the output sound based on the reference frequency band, the amplification factor, and the extracted phonemes. The adjustment unit 155 adjusts the output sound by multiplying the sound pressure of the reference frequency band by the amplification factor for each phoneme. As described above, the reference frequency band includes a promotion frequency band and a suppression frequency band.

[0066] The adjustment unit 155 adjusts the sound pressure of the reference frequency band using an amplification factor according to the phoneme extracted from the input voice. The amplification factor of the promotion frequency band is set to be greater than 1, and the amplification factor of the suppression frequency band is set to be less than 1. Therefore, the adjustment unit 155 adjusts the sound pressure of the output voice by increasing the sound pressure of the promotion frequency band and decreasing the sound pressure of the suppression frequency band. The adjustment unit 155 may adjust the sound pressure of either the promotion frequency band or the suppression frequency band. The adjustment unit 155 outputs the adjusted voice from the speaker 16.

[0067] Furthermore, when the amplification factor is updated, the adjustment unit 155 adjusts the output sound using the updated amplification factor. Specifically, the adjustment unit 155 adjusts the output sound based on the reference frequency band, the updated amplification factor, and the extracted phoneme. The adjustment unit 155 may cause the amplification factor determination unit 153 to update the amplification factor when the power supply of the hearing device 10 is turned on, and adjust the output sound using the updated amplification factor. Furthermore, the adjustment unit 155 may refer to the fourth memory M4 when the power supply of the hearing device 10 is turned on, and perform adjustment that reflects the adjustment result at the time of the previous use.

[0068] Furthermore, the adjustment unit 155 may adjust the output sound by taking into account environmental sounds contained in the sound. The processing unit 15 adjusts the sound pressure of the output sound according to the presence or absence of environmental sounds, the volume of the environmental sounds, or a user selection. For example, when emphasizing environmental sounds, the processing unit 15 increases the sound pressure of the frequency band corresponding to the environmental sounds. Furthermore, when reducing environmental sounds, the processing unit 15 decreases the sound pressure of the frequency band corresponding to the environmental sounds. Furthermore, the adjustment unit 155 may adjust the output sound based on a notification regarding the language received from the analysis unit 12.

[0069] Returning to Fig. 3, the speaker 16 is an example of an audio output unit that outputs audio. The speaker 16 outputs audio that has been adjusted by the adjustment unit 155.

[0070] The first memory M1 to the fourth memory M4 store various information according to this embodiment. The first memory M1 to the fourth memory M4 may be composed of, for example, a semiconductor memory. At least a part of the first memory M1 to the fourth memory M4 is composed of a non-volatile memory so that data is retained even when the power of the hearing device 10 is turned off.

[0071] The first memory M1 stores a specific frequency band. The specific frequency band is a frequency band in which the accuracy of recognizing a user's words is insufficient, regardless of the phonemes contained in the voice. The second memory M2 stores a reference frequency band. The reference frequency band includes a promotion frequency band in which the accuracy of recognizing a word is improved by increasing the sound pressure, and a suppression frequency band in which the accuracy of recognizing a word is improved by decreasing the sound pressure. The reference frequency band may correspond to the phonemes of multiple foreign languages.

[0072] The third memory M3 stores gains of sound pressure corresponding to the reference frequency bands determined according to the user, the gains of the promotion frequency bands being greater than 1, and the gains of the suppression frequency bands being less than 1.

[0073] The fourth memory M4 stores the adjustment result of the output sound. The adjustment result includes, for example, information on the phonemes that the user could not hear. The hearing device 10 can use the adjustment result for adjustment the next time it is used.

[0074] (Server 20 processing) The process performed by the server 20 will be described with reference to Fig. 5. Fig. 5 is a flowchart showing the flow of the reference frequency band determination process performed by the server 20. In this embodiment, the reference frequency band includes both the promotion frequency band and the suppression frequency band.

[0075] Before performing step S1, the recognition rate calculation unit 21 (see FIG. 2) calculates a standard recognition rate for each phoneme in advance. The recognition rate calculation unit 21 has a plurality of subjects listen to unadjusted words whose sound pressure has not been adjusted. For example, the recognition rate calculation unit 21 has the subjects listen to words including one or more of d, b, in, o, l, a, r, p, h, g, ch, sh, t, f, th, and s, which are phonemes concentrated in the frequency band from 500 Hz to 4000 Hz. The recognition rate calculation unit 21 may have the subjects listen to a plurality of words. It is preferable that the plurality of words are words from the same region.

[0076] The number of subjects is, for example, 100 or more, but is not limited to this. For example, each of the subjects listens to unadjusted words using the hearing device 10, and inputs the word recognition results to the hearing device 10. The word recognition results indicate how the word was heard. Note that, since the way a word is heard differs depending on the environment, the subjects need to hear the word under the same environmental conditions.

[0077] The subject may input the recognition result by voice via the microphone 11, for example. The subject may also input the recognition result to the hearing device 10 by voice, text, or the like using a terminal device (not shown). The terminal device may be, for example, a PC or a smartphone. The recognition rate calculation unit 21 receives the recognition result from the hearing device 10 and judges whether the recognition result is correct or incorrect. The hearing device 10 or the subject's terminal device may judge whether the recognition result is correct or incorrect, and transmit the judgment result to the recognition rate calculation unit 21.

[0078] The recognition rate calculation unit 21 calculates the recognition rate of a word based on the number of subjects and the number of people who answered correctly. Specifically, the recognition rate calculation unit 21 calculates the recognition rate by dividing the number of people who answered correctly by the number of subjects, as shown in the following formula (1). Recognition rate = number of correct answers / number of subjects (1)

[0079] For example, when the number of subjects is 100, the recognition rate calculation unit 21 calculates the recognition rate by the following formula (2). Recognition rate = number of correct answers / 100 (2)

[0080] In this way, the recognition rate calculation unit 21 obtains the recognition rate of the word in an unadjusted state for each phoneme. In this way, the recognition rate calculation unit 21 can determine the standard recognition rate for each phoneme.

[0081] Next, a process of determining the promotion frequency band and the inhibition frequency band in the reference frequency band determination unit 22 will be described. First, the reference frequency band determination unit 22 selects a phoneme (S1). For example, the reference frequency band determination unit 22 selects one phoneme from the above-mentioned d, b, in, o, l, a, r, p, h, g, ch, sh, t, f, th, and s.

[0082] The processes S2 to S9 indicated by the dashed lines in the figure are processes for identifying the reference frequency band of the selected phoneme. The reference frequency band determination unit 22 performs the processes S2 to S9 for each phoneme. This allows the reference frequency band determination unit 22 to identify, for each phoneme, the reference frequency band that improves the word recognition rate when the sound pressure is increased or decreased.

[0083] Next, the reference frequency band determination unit 22 selects a frequency band (S2). For example, the reference frequency band determination unit 22 selects one of a plurality of frequency bands separated by 50 Hz intervals. Here, the selected frequency band is referred to as a "selected frequency band." The reference frequency band determination unit 22 determines whether the subject's word recognition accuracy is improved when the sound pressure of the selected frequency band is lowered or raised. Here, the recognition rate is used as the recognition accuracy. The reference frequency band determination unit 22 determines whether the selected frequency band is a promotion frequency band or a suppression frequency band based on the determination result.

[0084] The flow of determining the suppression frequency band will be described using steps S3 to S5. First, the reference frequency band determination unit 22 lowers the sound pressure of the selected frequency band (S3). The recognition rate calculation unit 21 has the test subject listen to the words with the lowered sound pressure, and calculates the recognition rate using the above-mentioned formula (1). The reference frequency band determination unit 22 determines whether the recognition rate has improved (S4). If the recognition rate has not improved (NO in S4), the process returns to step S2. If the recognition rate has improved (YES in S4), the reference frequency band determination unit 22 determines that the selected frequency band is the suppression frequency band, and records the determination result in the storage device of the server 20 (S5).

[0085] The flow of determining the promotion frequency band will be described using steps S6 to S8. First, the reference frequency band determination unit 22 increases the sound pressure of the selected frequency band (S6). The recognition rate calculation unit 21 has the test subject listen to the words with the increased sound pressure, and calculates the recognition rate using the above-mentioned formula (1). The reference frequency band determination unit 22 determines whether the recognition rate has improved (S7). If the recognition rate has not improved (NO in S7), the process returns to step S2. If the recognition rate has improved (YES in S7), the reference frequency band determination unit 22 determines that the selected frequency band is the promotion frequency band, and records the determination result in the storage device of the server 20 (S8).

[0086] Specifically, for example, suppose that the recognition rate of hard-to-hear words is improved by increasing the sound pressure of frequency band a. In this case, the reference frequency band determination unit 22 determines frequency band a as the promotion frequency band. Also, suppose that the recognition rate of hard-to-hear words is improved by decreasing the sound pressure of frequency band b. In this case, the reference frequency band determination unit 22 determines frequency band b as the suppression frequency band.

[0087] Here, the reference frequency band determination unit 22 determines the reference frequency band by determining whether or not the recognition rate improves by increasing or decreasing the sound pressure, but this is not limited to the above. The reference frequency band determination unit 22 may determine the reference frequency band by determining whether or not the recognition rate decreases by increasing or decreasing the sound pressure. For example, if the recognition rate decreases due to an increase in the sound pressure of the selected frequency band, the reference frequency band determination unit 22 may determine the selected frequency band as the suppression frequency band. Similarly, if the recognition rate decreases due to a decrease in the sound pressure of the selected frequency band, the reference frequency band determination unit 22 may determine the selected frequency band as the promotion frequency band.

[0088] Next, the reference frequency band determination unit 22 judges whether or not the processing of all frequency bands has been completed (S9). If it is judged that the processing of all frequency bands has not been completed (NO in S9), the reference frequency band determination unit 22 returns to step S2 and repeats the subsequent processes. If it is judged that the processing of all frequency bands has been completed (YES in S9), the reference frequency band determination unit 22 judges whether or not the processing of all phonemes has been completed (S10).

[0089] If it is determined that the processing of all phonemes has not been completed (NO in S10), the reference frequency band determination unit 22 returns to step S1 and repeats the subsequent processing. If it is determined that the processing of all phonemes has been completed (YES in S10), the reference frequency band determination unit 22 ends the reference frequency band determination processing. The server 20 transmits information on the promotion frequency band and the suppression frequency band determined by this processing to the hearing device 10. The hearing device 10 records the received information on the promotion frequency band and the suppression frequency band in the second memory M2. This allows the server 20 to provide the hearing device 10 with information on the promotion frequency band and the suppression frequency band that are the basis for voice adjustment, which are determined based on data obtained from multiple subjects.

[0090] (Setting process of hearing device 10) The setting process performed by the hearing device 10 will be described with reference to Fig. 6. Fig. 6 is a flow chart showing the flow of the setting process performed by the hearing device 10. The setting process includes an initial setting process and an amplification factor determination process. In the process of Fig. 5 described above, the reference frequency band was determined based on the data of multiple subjects, but in this process, the hearing of the user himself / herself is identified, so that adjustments can be made to suit the word recognition situation of the individual user.

[0091] In steps S11 to S15, the hearing device 10 performs an initial setting process. The initial setting process is a process for adjusting the output sound for frequency bands that are difficult for the user to hear, similar to a general hearing aid function that does not take into account the influence of phonemes.

[0092] First, the initial setting unit 151 (see FIG. 4) selects a frequency band (S11). Here again, the selected frequency band is referred to as a "selected frequency band." The initial setting unit 151 plays a predetermined test sound to the user and determines whether the user can hear the test sound (S12). For example, the initial setting unit 151 receives the user's response to the test sound via the hearing device 10 or the user's terminal device.

[0093] If the test sound is not heard (NO in S12), the initial setting unit 151 increases the sound pressure of the selected frequency band (S13) and returns to step S12 for further determination. If the test sound is heard (YES in S12), the initial setting unit 151 records a specific frequency band in the first memory M1 (S14). The specific frequency band indicates a frequency band in which the user's word recognition accuracy is insufficient, regardless of the phonemes contained in the word.

[0094] The initial setting unit 151 determines whether or not the processing of all frequency bands has been completed (S15). If the processing of all frequency bands has not been completed (NO in S15), the initial setting unit 151 returns to step S11 and repeats the subsequent processes. If the processing of all frequency bands has been completed (YES in S15), the initial setting unit 151 proceeds to step S16. Note that the hearing device 10 may be configured to perform the processes of step S16 and subsequent steps without performing the processes of steps S11 to S15.

[0095] In steps S16 to S21, the hearing device 10 performs an amplification factor determination process. The amplification factor determination process is a process for determining an amplification factor suitable for a user in order to improve the recognition rate for each phoneme of the user. The amplification factor indicates the degree to which the sound pressure of the reference frequency band is amplified.

[0096] First, the amplification factor determination unit 153 selects a phoneme (S16). Here, the selected phoneme is referred to as the "selected phoneme." The amplification factor determination unit 153 causes a word including the selected phoneme to be output from the speaker 16, and determines whether the user's word recognition accuracy has improved compared to normal times (S17). Here, "normal times" refers to the timing after a general hearing aid function has been realized by the processing of steps S11 to S15. If the hearing device 10 does not perform steps S11 to S15, the normal times may be a time when the user is not wearing the hearing device 10.

[0097] The user inputs the recognition result via the hearing device 10 or another terminal device. The amplification factor determination unit 153, for example, determines whether the input recognition result is correct. The amplification factor determination unit 153 may determine whether the recognition accuracy has improved by comparing with the user's normal recognition result. For example, when a word that is normally recognized incorrectly is recognized correctly, the amplification factor determination unit 153 determines that the recognition accuracy of the word has improved. The amplification factor determination unit 153 may receive a response from the user as to whether the word was easy to hear and make a determination.

[0098] If the recognition accuracy has not improved (NO in S17), the amplification factor determination unit 153 refers to the second memory M2, acquires the promotion frequency band of the selected phoneme, and increases the sound pressure of the promotion frequency band (S18). Also, the amplification factor determination unit 153 refers to the second memory M2, acquires the suppression frequency band of the selected phoneme, and decreases the sound pressure of the suppression frequency band (S19). The amplification factor determination unit 153 returns to step S17, and repeats the processes of steps S17 and S18 until the recognition accuracy improves.

[0099] If the recognition accuracy has improved (YES in S17), the amplification factor determination unit 153 determines an amplification factor and records the amplification factor in the third memory M3 (S20). The amplification factor determination unit 153 determines whether or not the processing of all phonemes has been completed (S21). If the processing of all phonemes has not been completed (NO in S21), the amplification factor determination unit 153 returns to step S16 and repeats the subsequent processing. If the processing of all phonemes has been completed (YES in S21), the amplification factor determination unit 153 ends this processing.

[0100] (Audio Adjustment Processing of Hearing Device 10) The sound adjustment process performed by the hearing device 10 will be described with reference to Fig. 7. Fig. 7 is a flow chart showing the flow of the sound adjustment process performed by the hearing device 10. Here, the operation of the hearing device 10 operating as a hearing aid and adjusting the output sound will be described. The hearing device 10 may be worn on each of the left and right ears. The left and right hearing devices 10 may each perform the same process, or part of the process may be performed by either the left or right.

[0101] First, the microphone 11 (see FIG. 3) acquires sounds around the user (S31). The analysis unit 12 analyzes the sounds input from the microphone 11 and extracts phonemes (S32). The adjustment unit 155 (see FIG. 4) increases the sound pressure of the specific frequency band acquired from the first memory M1 (S33). This makes it possible to increase the sound pressure of the specific frequency band in which the user has poor hearing, thereby enabling the user to improve hearing of sounds in that frequency band. Through this process, the hearing device 10 realizes a general hearing aid function provided in well-known hearing aids.

[0102] Next, the adjustment unit 155 adjusts the output sound based on the reference frequency band acquired from the second memory M2, the amplification factor acquired from the third memory M3, and the phonemes extracted by the analysis unit 12 (S34). Specifically, the adjustment unit 155 adjusts the sound pressure of the reference frequency band using the amplification factor according to the extracted phonemes. The adjustment unit 155 increases the sound pressure of the promotion frequency band according to the amplification factor. Also, the adjustment unit 155 decreases the sound pressure of the suppression frequency band according to the amplification factor. The adjustment unit 155 outputs the adjusted sound from the speaker 16 (S35).

[0103] The AI13 judges whether the user can easily hear the voice (S36). The AI13 makes the judgment by acquiring the user's judgment, for example, via the microphone 11 or the user's terminal device. Alternatively, the AI13 may acquire the user's judgment by analyzing the user's voice and make the judgment. For example, if the user's voice contains repetition of the same word or a request to the speaker to repeat the word, the user may not have been able to hear the word. In this case, the AI13 judges that the user has difficulty hearing the voice. Conversely, if the user is having a smooth conversation, the AI13 judges that the user has easy hearing the voice.

[0104] If the user judges that the voice is not easy to hear (NO in S36), the adjustment unit 155 returns to step S34 and repeats the adjustment process until the voice is easy to hear. If the user judges that the voice is easy to hear (YES in S36), the adjustment unit 155 records the adjustment result in the fourth memory M4 (S37). For example, the adjustment unit 155 records data of words (phonemes) that the user could not hear in the fourth memory M4. Note that, if the user judges that the voice is not easy to hear in step S36 (NO in S36), the adjustment unit 155 may not perform the adjustment process, but may record the words (phonemes) that were difficult to hear in the fourth memory M4, and perform the adjustment process later. At this time, it is preferable to record time information and surrounding environmental information.

[0105] The adjustment unit 155 determines whether the adjustment is complete (S38). If it is determined that the adjustment is not complete (NO in S38), the hearing device 10 returns to step S31 and repeats the subsequent processes. If it is determined that the adjustment is complete (YES in S38), the hearing device 10 ends the sound adjustment process.

[0106] Fig. 8 and Fig. 9 are diagrams showing changes in frequency characteristics before and after adjustment of the output sound. Fig. 8 is a diagram showing frequency characteristics D1 before adjustment. Fig. 9 is a diagram showing frequency characteristics D2 after adjustment. In Fig. 8 and Fig. 9, the horizontal axis indicates frequency, and the vertical axis indicates sound pressure. Fig. 8 and Fig. 9 also show a promotion frequency band A and a suppression frequency band B.

[0107] The frequency characteristic D1 is, for example, the frequency characteristic of the sound input from the microphone 11. Alternatively, the frequency characteristic D1 may be the frequency characteristic of the sound after adjustment of a specific frequency band has been performed. The frequency characteristic D2 is the frequency characteristic of the output sound after adjustment based on the reference frequency band, the amplification factor, and the extracted phonemes.

[0108] In frequency characteristic D2 in Fig. 9, frequency characteristic D1 before adjustment is indicated by a thin solid line. As shown in Fig. 9, in frequency characteristic D2 after adjustment, the sound pressure in promotion frequency band A is higher than that in frequency characteristic D1 before adjustment. Also, in frequency characteristic D2, the sound pressure in suppression frequency band B is lower than that in frequency characteristic D1. By listening to the voice adjusted in this way, the user can easily hear words.

[0109] The configuration and processing of the system 100 according to this embodiment have been described above. The configuration of the system 100 described above is merely an example and may be modified as appropriate. For example, in the above description, an example was shown in which various information is recorded in the first memory M1 to the fourth memory M4, but this is not limiting. The hearing device 10 may obtain various information from other devices via the network N.

[0110] In the above description, the reference frequency band is determined by the server 20, but this is not limiting. The hearing device 10 may have the functionality of the server 20, and the hearing device 10 may determine the reference frequency band.

[0111] In the above description, the first memory M1 stores information on a specific frequency band where the user cannot hear well, regardless of the phoneme, but this configuration is not essential. For example, a hearing device 10 that does not have the function of such a general hearing aid and adjusts the output sound based on the phonemes contained in the sound may be realized. In this way, the present disclosure can be applied to cases where a person with normal hearing who has a sufficient hearing level in a general environment wants to adjust the sound he / she hears in a noisy environment.

[0112] As described above, in the system 100 according to the present embodiment, the server 20 determines a reference frequency band for each phoneme, including a promotion frequency band for improving word recognition accuracy by increasing sound pressure and a suppression frequency band for improving recognition accuracy by decreasing sound pressure. The hearing device 10 acquires information on the reference frequency band from the server 20. The hearing device 10 also determines an amplification factor of the sound pressure corresponding to the reference frequency band according to the user. The hearing device 10 analyzes the sound around the user to extract phonemes contained in the sound, and adjusts the output sound based on the reference frequency band, the amplification factor, and the extracted phonemes.

[0113] With this configuration, the hearing device 10 can adjust the output voice appropriately according to the phonemes contained in the input voice, thereby improving the word recognition rate of the user. The hearing device 10 not only increases the sound pressure of a specific frequency band, but also adjusts the sound pressure taking into account the influence of the phonemes, so that, for example, in the case where the user is hearing-impaired, the hearing aid effect can be improved.

[0114] Furthermore, the hearing device 10 can update the amplification factor according to the user's word recognition status. This allows the hearing device 10 to adjust the output sound using the updated amplification factor even if the user has difficulty hearing a word in a conversation.

[0115] Furthermore, the hearing device 10 can analyze the voice and extract phonemes while taking into account the environmental sounds contained in the voice, so that the output voice can be appropriately adjusted even when the user moves to a different environment. For example, even when the user is having a conversation in a place with a loud environmental sound, the user can easily recognize the speaker's words by listening to the output voice adjusted by the hearing device 10. It is preferable that the hearing device 10 reduces the amount of calculations and reduces processing time and power consumption by keeping the adjustment results and fine-tuning the past adjustment results in a similar environment.

[0116] Each of the functional components of the above-mentioned hearing device 10 and server 20 may be realized by hardware that realizes each functional component (e.g., hardwired electronic circuitry, etc.), or may be realized by a combination of hardware and software (e.g., a combination of an electronic circuitry and a program that controls it, etc.). For example, any process in the present disclosure may be realized by having a CPU execute a computer program.

[0117] The program includes instructions (or software code) that, when loaded into a computer, cause the computer to perform one or more functions described in the embodiments. The program may be stored on various types of non-transitory computer readable medium or tangible storage medium. By way of example and not limitation, non-transitory computer readable medium or tangible storage medium includes random-access memory (RAM), read-only memory (ROM), flash memory, solid-state drive (SSD) or other memory technology, CD-ROM, digital versatile disc (DVD), Blu-ray® disk or other optical disk storage, magnetic cassette, magnetic tape, magnetic disk storage or other magnetic storage device. The program may also be transmitted on various types of transitory computer readable medium or communication medium. By way of example and not limitation, transitory computer readable medium or communication medium includes electrical, optical, acoustic, or other form of propagated signal.

[0118] The present disclosure is not limited to the above-described embodiment, and can be modified as appropriate without departing from the spirit and scope of the present disclosure.

[0119] For example, the microphone 11 or speaker 16 of the hearing device 10 may be separated, and signal processing may be performed by a separate device such as a smartphone. In this configuration, the battery of the smartphone may be used, thereby increasing the continuous operating time of the hearing device 10.

[0120] Additionally, the hearing device 10 may transmit data relating to the results of the adjustment to the server 20. In this manner, the server 20 can collect and utilize data from a large number of users as big data.

[0121] Furthermore, the division of roles between the hearing device 10 and the server 20 can be changed as appropriate. Some of the functions of the server 20 may be executed by the hearing device 10, or some of the functions of the hearing device 10 may be executed by the server 20. This can be changed as appropriate as necessary. [Explanation of symbols]

[0122] 10 Listening Device 11. Mike 12 Analysis Department 13 AI 14 Communications Department 15 Processing section 16 Speakers 20 Servers 21 Recognition rate calculation section 22 Reference frequency band determination unit 24 Communications Department 100 Systems 151 Initial setting section 152 Reference frequency band acquisition unit 153 Amplification factor determination unit 154 Amplification factor acquisition unit 155 Adjustment section A Promotion Frequency Band B Suppression frequency band D1 Frequency Response D2 frequency characteristics M1 The first memory M2 Second Memory M3 The third memory M4 The fourth memory N Network

Claims

1. a reference frequency band acquisition unit that acquires, for each phoneme, a reference frequency band including at least one of a first frequency band for improving word recognition accuracy by increasing sound pressure and a second frequency band for improving the recognition accuracy by decreasing sound pressure; an amplification factor acquisition unit that acquires an amplification factor of a sound pressure corresponding to the reference frequency band determined according to a user; A voice input unit for inputting voices around the user; an analysis unit that analyzes the speech and extracts phonemes contained in the speech; an adjustment unit that adjusts an output voice based on the reference frequency band, the amplification factor, and the extracted phonemes. Listening device.

2. An update unit that updates the amplification factor in response to a word recognition situation of the user, The adjustment unit adjusts the output sound using the updated amplification factor.

2. A hearing device according to claim 1.

3. The analysis unit analyzes the voice while taking into account environmental sounds contained in the voice, and extracts the phonemes. A hearing device according to claim 1 or 2.

4. The computer a reference frequency band acquisition step of acquiring, for each phoneme, a reference frequency band including at least one of a first frequency band for improving word recognition accuracy by increasing sound pressure and a second frequency band for improving the recognition accuracy by decreasing sound pressure; an amplification factor acquisition step of acquiring an amplification factor of a sound pressure corresponding to the reference frequency band determined according to a user; a voice input step of inputting voice around the user; an analysis step of analyzing the speech and extracting phonemes contained in the speech; and adjusting an output voice based on the reference frequency band, the amplification factor, and the extracted phonemes. Listening methods.

5. a reference frequency band acquisition step of acquiring, for each phoneme, a reference frequency band including at least one of a first frequency band for improving word recognition accuracy by increasing sound pressure and a second frequency band for improving the recognition accuracy by decreasing sound pressure; an amplification factor acquisition step of acquiring an amplification factor of a sound pressure corresponding to the reference frequency band determined according to a user; a voice input step of inputting voice around the user; an analysis step of analyzing the speech and extracting phonemes contained in the speech; and adjusting an output voice based on the reference frequency band, the amplification factor, and the extracted phonemes. program.

Images