Control method of electronic device providing auditory temporal resolution rehabilitation training

Abstract

A method of operation of an electronic device for performing auditory temporal resolution training to rehabilitate a user's speech recognition ability is disclosed. A method of operation of an electronic device according to the present disclosure comprises: a first inspection step of providing a user with a pair of audio data comprising a first audio data that matches a first sound and a second audio data that matches a second sound in which at least a portion of the first sound is replaced with a blank space; a second inspection step of increasing the blank space interval of the second audio data and providing a pair of audio data including the first audio data and the second audio data with the increased blank space interval when a user input corresponding to an incorrect answer selecting the first audio data in the first inspection step is received and a third inspection step of decreasing the blank space interval of the second audio data and providing a pair of audio data including the first audio data and the second audio data with the decreased blank space interval when a user input corresponding to an correct answer selecting the second audio data in the first inspection step is received and providing a pair of audio data including the first audio data and the second audio data with the decreased blank space interval.

Description

Technology Field

[0001] The present disclosure relates to a method of operation of an electronic device, and more specifically, to a method of operation of an electronic device for auditory temporal resolution training, wherein the method comprises providing a user with a pair of audio data including audio data classified according to stimulation frequency, stimulation intensity, stimulation length, etc., and audio data in which a portion of the said audio data is replaced with a blank interval, wherein the length of the blank interval is changed according to the user's selection to obtain a gap detection threshold value, which is a minimum length blank interval identifiable by the user. Background Technology

[0002] The process of hearing involves complex auditory elements being converted from sound waves into electrophysiological signals and transmitted from the periphery to the center. Auditory elements include the frequency, loudness, time, length, speed, and sound quality of the sound. When the complex characteristics of each element are transmitted as electrical nerve signals starting from the outer ear, a peripheral auditory organ, to the auditory cortex, a central organ, the input sound elements are analyzed and processed to enable the perception of sound; this ability is called central auditory processing (CAP).

[0003] Auditory temporal processing, which handles the temporal characteristics of sound within central auditory processing, is the ability to process presented sounds rapidly in time and is divided into four functions: temporal ordering, temporal integration, temporal masking, and temporal resolution. Among these, auditory temporal resolution (ATR) is the ability to discriminate and perceive subtle temporal differences in sound; in particular, it plays a crucial role in speech discrimination by providing clues that allow one to identify the boundaries between syllables or words. Therefore, a decline in auditory temporal resolution can cause distortion in sound perception, which ultimately leads to distorted speech and results in impaired speech recognition.

[0004] Auditory temporal resolution is associated with neural processing speed, and the decline in neural processing speed due to aging can affect the temporal resolution of the elderly, making speech recognition difficult. This decline in neural processing speed in the elderly can be caused by various factors, including delayed neural recovery time, degeneration of neural connections between brain regions, damage to brain mechanisms, loss of neural synchronization, and loss of myelination. Furthermore, hearing loss can also affect auditory temporal resolution, and individuals with hearing impairment exhibit a more significant decline in this ability. This suggests that most elderly individuals with hearing loss may experience a decline in auditory temporal resolution.

[0005] As of 2023, the elderly population in Korea accounted for 18.4% of the total population and is projected to continue increasing to 20.6% by 2025, entering a super-aged society. Considering that more than one-third of the current elderly population suffers from hearing loss that causes functional defects, there is a need for research on the diagnosis, prevention, and rehabilitation of the decline in auditory temporal resolution in the elderly as the number of the elderly and hearing-impaired population gradually increases.

[0006] However, while technological advancements such as hearing aids and cochlear implants continue to help address the decline in hearing ability among the elderly, there is a lack of domestic auditory training technologies aimed at directly improving the central auditory nervous system. In particular, training tools specialized for central auditory processing have not been developed domestically, and auditory temporal processing training programs are also absent. Therefore, there is a need to develop auditory temporal resolution rehabilitation training programs for the elderly in Korea. Prior art literature

[0007] Registered Patent Publication No. 10-2556571 The problem to be solved

[0008] The purpose of the present disclosure is to provide an electronic device capable of providing multiple audio contents according to the frequency, length, and intensity of the stimulus sounds, which are characteristics of auditory temporal resolution training content.

[0009] In addition, the purpose of this disclosure is to set training content suitable for each user based on the performance level of multiple training subjects.

[0010] The purposes of the present disclosure are not limited to those mentioned above, and other purposes and advantages of the present disclosure not mentioned may be understood from the following description and will be more clearly understood from the embodiments of the present disclosure. Furthermore, it will be readily apparent that the purposes and advantages of the present disclosure can be realized by the means and combinations thereof set forth in the claims. means of solving the problem

[0011] A method of operation of an electronic device for performing auditory temporal resolution training for the rehabilitation of a user’s speech recognition ability according to one embodiment of the present disclosure comprises: a first inspection step of providing a user with a pair of audio data consisting of a first audio data matched to a first sound and a second audio data matched to a second sound in which at least a portion of the first sound is replaced with a blank space; a second inspection step of increasing the blank space interval of the second audio data and providing the user with a pair of audio data including the first audio data and the second audio data with the increased blank space interval when user input corresponding to an incorrect answer of selecting the first audio data in the first inspection step is received and providing the user with a pair of audio data including the first audio data and the second audio data with the reduced blank space interval when user input corresponding to an correct answer of selecting the second audio data in the first inspection step is received and providing the user with a pair of audio data including the first audio data and the second audio data with the reduced blank space interval.

[0012] At this time, the method of operation of the electronic device may include a step of performing a plurality of inspection steps while repeating a process of changing the gap interval of the second audio data according to a user input selecting one of the first audio data and the second audio data, starting from the second inspection step, and a step of obtaining a gap detection threshold value, which is a minimum length gap interval identifiable by the user, based on the gap interval of the second audio data at the point in time when the user input selecting one of the first audio data and the second audio data is inverted a certain number of times through the plurality of inspection steps.

[0013] In this case, the step of obtaining the interval detection threshold value may obtain the user's interval detection threshold value according to the average value of the gap interval of the second audio data identified for each of the most recent preset number of inversion points based on the point in time that has been inverted a certain number of times.

[0014] Meanwhile, the method of operation of the electronic device may include a step of performing a plurality of inspection steps while repeating a process of changing the gap interval of the second audio data according to a user input selecting one of the first audio data and the second audio data, starting from the third inspection step, and a step of obtaining a gap detection threshold value, which is a minimum length gap interval identifiable by the user, based on the gap interval of the second audio data at the point in time when the user input selecting one of the first audio data and the second audio data is inverted a certain number of times through the plurality of inspection steps.

[0015] Additionally, the first inspection step may receive a user input selecting one of the first audio data and the second audio data by providing the first audio data and the second audio data in any order, and the method of operation of the electronic device may perform the second inspection step of increasing the gap of the second audio data when the user input corresponding to the incorrect answer of selecting the first audio data is received in the first inspection step, and may perform the first inspection step again when the user input corresponding to the correct answer of selecting the second audio data is received in the first inspection step, and may perform the second inspection step of increasing the gap of the second audio data when the user input corresponding to the incorrect answer of selecting the first audio data is received in the first inspection step performed again, and may perform the third inspection step of decreasing the gap of the second audio data when the user input corresponding to the correct answer of selecting the second audio data is received in the first inspection step performed again.

[0016] Additionally, the method of operating the electronic device may include the step of generating a plurality of first audio data in which at least one of the stimulation frequency and the stimulation length is different based on a plurality of stimulation frequencies and a plurality of stimulation lengths, and the method of operating the electronic device may perform at least one of the first inspection step, the second inspection step, and the third inspection step based on each of the plurality of first audio data.

[0017] Meanwhile, the second inspection step may increase the gap interval of the second audio data by a first threshold value, and the third inspection step may decrease the gap interval of the second audio data by a second threshold value. The method of operation of the electronic device may include a step of setting an expected gap detection threshold value of the user based on a gap detection threshold value obtained for each of the plurality of users, and a step of setting the first threshold value and the second threshold value whenever a plurality of inspection steps are performed based on the set expected gap detection threshold value. In the step of setting the first threshold value and the second threshold value, the first threshold value and the second threshold value may be set to a larger value as the difference between the gap interval of the second audio data and the expected gap detection threshold value is greater, and the first threshold value and the second threshold value may be set to a smaller value as the difference between the gap interval of the second audio data and the expected gap detection threshold value is smaller.

[0018] An electronic device for performing auditory temporal resolution training for the rehabilitation of a user's speech recognition ability according to one embodiment of the present disclosure comprises: a memory storing a plurality of sounds for auditory temporal resolution training; a first test step of providing a user with a pair of audio data comprising a first audio data matching a first sound and a second audio data matching a second sound in which at least a portion of the first sound is replaced with a blank space; a second test step of providing a user with a pair of audio data including the first audio data and the second audio data with the increased blank space when user input corresponding to an incorrect answer of selecting the first audio data is received in the first test step; and a third test step of providing a user with a pair of audio data including the first audio data and the second audio data with the increased blank space when user input corresponding to an correct answer of selecting the second audio data is received in the first test step.

[0019] An auditory temporal resolution training system for the rehabilitation of a user's speech recognition ability according to one embodiment of the present disclosure comprises: an electronic device that performs a first inspection step of providing a pair of audio data to a user terminal, the first audio data matching a first sound and a second audio data matching a second sound in which at least a portion of the first sound is replaced with a blank space; a second inspection step of increasing the blank space interval of the second audio data and providing a pair of audio data including the first audio data and the second audio data with the increased blank space interval to the user terminal when a user input corresponding to an incorrect answer selecting the first audio data is received from the user terminal in the first inspection step; and a third inspection step of decreasing the blank space interval of the second audio data and providing a pair of audio data including the first audio data and the second audio data with the decreased blank space interval to the user terminal when a user input corresponding to a correct answer selecting the second audio data is received from the user terminal in the first inspection step; and a user terminal that receives the pair of audio data from the electronic device and provides a user input selecting the audio data to the electronic device.

[0020] The present disclosure includes a non-transient computer-readable medium storing at least one instruction that is executed by a processor of an electronic device to cause said electronic device to perform the method of operation of said electronic device described above. Effects of the invention

[0021] Through the present disclosure, the user's auditory temporal resolution ability can be measured through audio content provided to the user, and the difference in interval detection threshold values ​​according to stimulus frequency, length, and intensity for each user can be effectively identified as a result of the measurement. Brief explanation of the drawing

[0022] FIG. 1 is a drawing for explaining the configuration of an electronic device according to one embodiment of the present disclosure, FIG. 2 is a drawing for explaining the operation of an electronic device according to one embodiment of the present disclosure, FIG. 3 is an algorithm for explaining the operation of an electronic device according to one embodiment of the present disclosure performing training according to adaptive learning, FIG. 4 is a diagram illustrating the operation of an electronic device according to one embodiment of the present disclosure to calculate a user's gap detection threshold value as a result of training being performed. FIG. 5 is a diagram illustrating the results of a re-examination conducted to verify consistency with the test results two weeks after an electronic device according to one embodiment of the present disclosure performed first training on 10 hearing adults. FIG. 6 is a diagram for comparing the training results of adults and elderly people for each frequency condition after auditory temporal resolution training of an electronic device according to one embodiment of the present disclosure has been performed. FIG. 7 is a diagram for comparing the individual ATR thresholds of an elderly person with the average value of a healthy adult as a result of auditory temporal resolution training of an electronic device according to one embodiment of the present disclosure. FIG. 8 is a drawing for explaining an auditory time resolution training system according to one embodiment of the present disclosure. Specific details for implementing the invention

[0023] Before specifically describing the present disclosure, the method of description in the specification and drawings is described.

[0024] First, the terms used in this specification and claims have been selected based on general terms considering their functions in the various embodiments of this disclosure. However, these terms may vary depending on the intent of those skilled in the art, legal or technical interpretations, and the emergence of new technologies. Additionally, some terms have been arbitrarily selected by the applicant. Such terms may be interpreted according to the meanings defined in this specification; in the absence of specific definitions, they may be interpreted based on the overall content of this specification and common technical knowledge in the relevant field.

[0025] In addition, the same reference numbers or symbols described in each drawing attached to this specification represent parts or components that perform substantially the same function. For convenience of explanation and understanding, the same reference numbers or symbols are used to describe different embodiments. That is, even if components having the same reference number are all depicted in multiple drawings, the multiple drawings do not imply a single embodiment.

[0026] Additionally, in this specification and claims, terms including ordinal numbers, such as "first," "second," etc., may be used to distinguish between components. These ordinal numbers are used to distinguish identical or similar components from one another, and the meaning of the terms should not be limited by the use of such ordinal numbers. For example, the order of use or arrangement of components combined with such ordinal numbers should not be restricted by the number. If necessary, each ordinal number may be used interchangeably.

[0027] In this specification, singular expressions include plural expressions unless the context clearly indicates otherwise. In this application, terms such as "comprising" or "consisting of" are intended to specify the existence of the features, numbers, steps, actions, components, parts, or combinations thereof described in the specification, and should be understood as not precluding the existence or addition of one or more other features, numbers, steps, actions, components, parts, or combinations thereof.

[0028] In the embodiments of the present disclosure, terms such as "module," "unit," "part," etc. are used to refer to a component that performs at least one function or operation, and such component may be implemented in hardware or software, or in a combination of hardware and software. Additionally, a plurality of "modules," "units," "parts," etc. may be integrated into at least one module or chip and implemented as at least one processor, except where each needs to be implemented in specific individual hardware.

[0029] Furthermore, in the embodiments of the present disclosure, when a part is described as being connected to another part, this includes not only a direct connection but also an indirect connection through another medium. Additionally, the meaning that a part includes a certain component implies that, unless specifically stated otherwise, it does not exclude other components but may include additional components.

[0030] Hereinafter, various embodiments of the present disclosure will be described in detail with reference to the attached drawings.

[0031] FIG. 1 is a drawing for explaining the configuration of an electronic device (100) according to one embodiment of the present disclosure.

[0032] According to FIG. 1, the electronic device (100) may include a memory (110), an audio output unit (120), a display (130), a user input unit (140), and a processor (150).

[0033] The electronic device (100) may be implemented as an information providing device for identifying minute temporal differences in sound between audio data provided to the user. Specifically, the electronic device (100) may provide information related to a minimum length gap interval that the user can identify within the audio data by providing multiple audio data. Additionally, the electronic device (100) may correspond to various terminal devices such as a desktop PC, laptop PC, tablet PC, smartphone, etc., capable of running software and / or applications for obtaining a minimum length gap interval that the user can identify, but is not limited thereto.

[0034] The memory (110) is configured to store at least one instruction or data related to an operating system (OS) for controlling the overall operation of the components of the electronic device (100) and the components of the electronic device (100).

[0035] The memory (110) may include non-volatile memory such as ROM or flash memory, and may include volatile memory such as DRAM. Additionally, the memory (110) may include a hard disk, an SSD (Solid state drive), etc.

[0036] The memory (110) may contain information about at least one task regarding the operation of the electronic device (100). In this case, the task may refer to an overall task for performing auditory temporal resolution training.

[0037] The audio output unit (120) is configured to output audio data and may include a speaker, an earphone terminal, etc. In one embodiment, the electronic device (100) may provide audio data for auditory time resolution testing to a user through a headphone device connected to the earphone terminal of the audio output unit (120).

[0038] The display (130) is configured to visually provide various information and can be implemented using LED, OLED, PDP, Micro LED, etc., and may include a touch panel for receiving user input (e.g., touch).

[0039] In one embodiment, the electronic device (100) may provide a user with a UI (User Interface) that allows the user to select audio data through a display (130).

[0040] The user input unit (140) is configured to receive various commands or information from the user. The user input unit (140) may be implemented with at least one button, touch pad, touch screen, microphone, sensor, etc. In this case, the touch pad may be combined with the display (130) to receive user input. Additionally, the electronic device (100) may be connected to various user input devices (e.g., keyboard, mouse, etc.) equipped with at least one keypad, button, motion sensor, etc.

[0041] For example, through a user input unit (140) such as a button, the electronic device (100) can obtain a user input that selects one of the audio data output to the audio output unit (120).

[0042] A processor (150) is configured to control the electronic device (100) overall. Specifically, the processor (150) can perform operations according to various embodiments of the present disclosure by being connected to memory (110) and executing at least one instruction stored in memory (110).

[0043] In particular, the processor (150) can be implemented as a single processor, as well as as multiple processors.

[0044] The processor (150) may be implemented in various ways. For example, one or more processors may include one or more of a CPU (Central Processing Unit), GPU (Graphics Processing Unit), APU (Accelerated Processing Unit), MIC (Many Integrated Core), DSP (Digital Signal Processor), NPU (Neural Processing Unit), hardware accelerator, or machine learning accelerator. One or more processors may control one or any combination of other components of the server and may perform operations or data processing related to communication. One or more processors may execute one or more programs or instructions stored in memory (110). For example, one or more processors may perform a method according to one embodiment of the present disclosure by executing one or more instructions stored in memory (110).

[0045] When a method according to one embodiment of the present disclosure includes a plurality of operations, the plurality of operations may be performed by a single processor or by a plurality of processors. For example, when a first operation, a second operation, and a third operation are performed by a method according to one embodiment, the first operation, the second operation, and the third operation may all be performed by a first processor, or the first operation and the second operation may be performed by a first processor (e.g., a general-purpose processor) and the third operation may be performed by a second processor (e.g., an artificial intelligence dedicated processor).

[0046] One or more processors may be implemented as a single-core processor comprising one core, or as one or more multicore processors comprising multiple cores (e.g., homogeneous multicore or heterogeneous multicore). When one or more processors are implemented as multicore processors, each of the multiple cores included in the multicore processor may include internal processor memory such as on-chip memory, and a common cache shared by the multiple cores may be included in the multicore processor. Additionally, each of the multiple cores included in the multicore processor (or some of the multiple cores) may independently read and execute program instructions for implementing the method according to one embodiment of the present disclosure, or all (or some) of the multiple cores may be linked together to read and execute program instructions for implementing the method according to one embodiment of the present disclosure.

[0047] When a method according to one embodiment of the present disclosure includes a plurality of operations, the plurality of operations may be performed by one of the plurality of cores included in a multi-core processor, or may be performed by a plurality of cores. For example, when a first operation, a second operation, and a third operation are performed by a method according to one embodiment, the first operation, the second operation, and the third operation may all be performed by a first core included in a multi-core processor, or the first operation and the second operation may be performed by a first core included in a multi-core processor and the third operation may be performed by a second core included in a multi-core processor.

[0048] In embodiments of the present disclosure, the processor (150) may mean a system-on-chip (SoC) in which one or more processors and other electronic components are integrated, a single-core processor, a multi-core processor, or a core included in a single-core processor or a multi-core processor, wherein the core may be implemented as a CPU, GPU, APU, MIC, DSP, NPU, hardware accelerator or machine learning accelerator, etc., but the embodiments of the present disclosure are not limited thereto.

[0049] Meanwhile, although not illustrated in FIG. 1, the electronic device (100) may include a communication unit for communicating with at least one external device.

[0050] The communication unit is configured for the electronic device (100) to communicate with an external device, such as a user terminal.

[0051] The communication unit may include circuits, modules, chips, etc., for communicating with external devices using various wired or wireless communication methods. The communication unit may also be connected to external devices through various networks.

[0052] Depending on the area or scale, a network may be a Personal Area Network (PAN), Local Area Network (LAN), Wide Area Network (WAN), etc., and depending on the openness of the network, it may be an Intranet, Extranet, or Internet, etc.

[0053] The communication unit can be connected to external devices through various wireless communication methods such as LTE (long-term evolution), LTE-A (LTE Advance), 5G (5th Generation) mobile communication, CDMA (code division multiple access), WCDMA (wideband CDMA), UMTS (universal mobile telecommunications system), WiBro (Wireless Broadband), GSM (Global System for Mobile Communications), DMA (Time Division Multiple Access), WiFi (Wi-Fi), WiFi Direct, Bluetooth, BLE (Bluetooth Low Energy), NFC (near field communication), Zigbee, and LoRa.

[0054] In addition, the communication unit may be connected to external devices via wired communication methods such as Ethernet, optical networks, USB (Universal Serial Bus), and Thunderbolt.

[0055] In addition, the communication department may be configured to utilize various newly devised communication methods / technologies in the future.

[0056] In one embodiment, the electronic device (100) may transmit audio data to an external device, such as a user terminal, through a communication unit instead of outputting audio data, and may obtain user input for selecting audio data instead of a user input unit (140).

[0057] Generally, auditory temporal resolution is measured by the ability to detect short gaps between sounds. A common method for evaluating gap detection ability involves having the user identify the presence or absence of a gap between specific presented sounds to determine the minimum detectable gap length (gap detection test); this is referred to as the gap detection threshold (GDT) or ATR threshold. There is a negative correlation between the GDT and auditory temporal resolution, meaning that a lower threshold indicates higher resolution. The gap detection threshold can be influenced by various factors, including top-down factors associated with cognitive abilities such as attention, and bottom-up factors such as stimulus characteristics and testing methods. Factors related to stimulus characteristics include the frequency, intensity, length, bandwidth, and location of the gap, while stimulus methods vary, such as fixed-type and randomized approaches.

[0058] The present disclosure investigated the effects on GDT by applying various of the aforementioned factors. When examining the changes in GDT according to factors, the first factor is frequency. The stimulus sound consisted of broad band noise (BBN) and narrow band noise (NBN) with various center frequencies ranging from 100 Hz to 6500 Hz; as the frequency increased, the threshold decreased (Fitzgibbons & Whiteman, 1982; Shailer, 1983; Shailer & Moore, 1985; Moore et al., 1993). Furthermore, regardless of the signal-to-noise ratio (SNR) between the stimulus sound and background noise, the threshold increased as the frequency decreased (Shailer, 1985). There were also differences between frequencies when measured at various intensities, with the highest threshold observed at the lowest frequency (Moore, 1993).

[0059] The second factor is intensity. As intensity decreased, the threshold also decreased (Plomp, 1964; Strouse, 1998), but depending on the specific intensity category, the variation in threshold was minimal or showed an increase of more than 50% (Moore et al., 1993). As SNR increased, the threshold decreased, but the difference was minimal at 12–15 dB (Shailer, 1985).

[0060] Thirdly, there is the change in GDT according to frequency bandwidth. Bandwidth is determined by the width relative to the center frequency or by the cut-off frequency filter. Bandwidth is related to the number of frequency channels; as the width increases, more frequencies are contained within the stimulus sound, which increases the frequency cues available to detect gaps and thus lowers the threshold. As bandwidth increases, the threshold decreases (Shailer & Moore, 1985; Snell et al., 1994), and this effect is more pronounced at low frequencies below 1000 Hz (Shailer, 1983). The reason threshold variation according to bandwidth is prominent at low frequencies can be attributed to psychoacoustic factors, specifically the auditory filter of the peripheral auditory system and frequency waveform characteristics. The auditory filter refers to the psychoacoustic range within which frequencies can be perceived, and it exhibits a narrower shape at low frequencies than at high frequencies (Moore & Glasberg, 1983). The narrower the range, the longer the stimulus duration required to perceive sound at low frequencies; however, as the gap becomes shorter, rapid temporal changes are not perceived, and the gap is heard as if it were filled with the stimulus sound (Park, 2011). Another reason is frequency fluctuation, which refers to the degree of the frequency waveform over time. Frequency occurs as a longitudinal wave that moves and causes pressure changes over time; as the bandwidth narrows, the frequency propagation speed slows down, making it difficult to perceive the interval between sounds. Furthermore, the reduced propagation speed decreases the transmitted energy, affecting gap detection (Shailer & Moore, 1983; Park, 2011).

[0061] In the case of stimulus length, generally, the threshold decreases as the stimulus length decreases. Schneider (1999) measured the threshold at stimulus lengths ranging from 2.5 ms to 500 ms and found that for both the elderly and adults, the threshold decreased from 500 ms to 5 ms and then increased again from 5 ms to 2.5 ms (Schneider, 1999). This change in threshold according to stimulus length can be explained by spectral splatter (Schneider et al., 1994; Suied et al., 2014). Spectral splatter refers to the phenomenon where energy remains in the center frequency region when the stimulus sound is sustained for a long time, but as the stimulus length decreases, energy is distributed to surrounding frequencies. When the range of frequencies where energy is distributed widens, there are more frequency channels that can provide clues for gap detection, making it easier to detect gaps using spectral changes rather than temporal changes as clues. To prevent this, studies use masking noise to suppress the audibility of the surrounding spectrum or use wide rise and fall times to reduce the spectral splatter effect.

[0062] The location of the gap can also affect threshold fluctuations; while the effect is minimal in adults, in the elderly, the threshold was lowest when the gap was located in the center of the stimulus sound and highest when it was located at the front. Regarding presentation methods, the threshold was higher when the stimulus sound was presented in a fixed position rather than randomly (Harris, 2010). As such, the elderly respond more sensitively to stimulus factors than adults, and among these factors, hearing is the most prominent. Although auditory temporal resolution ability due to hearing loss varies from person to person, it shows a higher threshold compared to adults with normal hearing. The elderly generally exhibit hearing loss at high frequencies above 4000 Hz (Lough, 2022). Experiments conducted in Korea to determine whether GDT in the elderly could be predicted based on hearing level showed that, based on 2 kHz, hearing in the high-frequency range had a greater impact on GDT than in the low-frequency range.

[0063] Meanwhile, the gap detection tests widely applied in actual clinical practice are the Gap in Noise (GIN) and the Random Gap Detection Test.

[0064] The GIN test involves presenting a subject with white noise of a comfortable intensity for 6 seconds and having them detect the gaps included within it. The track consists of one practice track and four actual test tracks, and the presentation is conducted at 50 dB SL based on the average of the subject's hearing thresholds at 0.5, 1, and 2 kHz. In each trial, white noise is presented for 6 seconds, and 0 to 3 gaps of varying lengths appear; the subject must press a button whenever a gap is identified. The gap lengths are 2, 3, 4, 5, 6, 8, 10, 12, 15, and 20 ms, from which three are presented randomly; however, each length is presented 6 times per track, resulting in a total of 60 gaps. Among the gap lengths, the gap that shows a correct response rate of 4 out of 6 (66.6%) is calculated as the approximate threshold (A.th.). If the correct response rate decreases after the first gap showing 66.6%, the length of the gap showing 66.6% thereafter is set as the proximity threshold. The total correct response rate is calculated by converting the number of correct responses out of a total of 60 gap presentations into a percentage (Musiek et al., 2005). When the GIN test was conducted on 50 hearing adults (Musiek, 2005), there was no difference in GDT between the two ears, with the right ear at 4.9 ms and the left ear at 4.8 ms. When tested on 18 subjects with deficits in central auditory processing, the right ear was 8.5 ms and the left ear was 7.8 ms; compared to hearing adults, subjects with deficits in central auditory processing had significantly lower GDTs. In the threshold results of 100 subjects with a larger number of participants (Giannela Samelli et al., 2008), the average GDT was 4.19 ms, which was close to the results of Musiek's study, with 3 ms being 10–30% and 4 ms being 60–70% according to the psychoacoustic function curve.

[0065] When GIN tests were conducted at 50 dB on 40 Korean adults with normal hearing (Choi et al., 2013), the results were 5.03 ms for men and 4.98 ms for women. Additionally, when GDT was measured at 20, 30, 40, and 50 dB SL on 10 adults with normal hearing, 12 elderly individuals with normal hearing, and 12 elderly individuals with hearing loss (Park & ​​Lee, 2016), the results were 7.5, 5.9, 4.8, and 4.2 ms for adults; 9.8, 8.1, 7.1, and 6.9 ms for elderly individuals with normal hearing; and 13.8, 11.3, 10, and 9.6 ms for elderly individuals with hearing loss. Significant differences in GDT were observed between age groups, with adults showing lower thresholds than the elderly, and elderly individuals with normal hearing showing lower thresholds than elderly individuals with hearing loss. Regarding intensity, regardless of age, GDT was measured to be lower at higher intensities.

[0066] While the GIN test uses BBN as the stimulus sound and does not provide information on frequency specificity, the RGDT test is capable of measuring gap detection thresholds for each frequency. This test presents stimulus sounds at short intervals at an intensity comfortable for the subject to hear, and involves identifying the short gaps included in between. Stimuli of 500, 1000, 2000, and 4000 Hz, along with clicks, are presented with short gaps of 17 ms in length. The gap lengths are 0, 2, 5, 10, 15, 20, 25, 30, and 40 ms, presented once at regular intervals in a random order, and the test proceeds with the subject pressing a button whenever a gap is detected. RGDT ability decreases with aging. As a result of conducting RGDT on hearing adults and hearing elderly individuals aged 50 to 67, at 500, 1000, 2000, and 4000 Hz, the GDT values ​​were 8.23, 8.62, 8.15, 6.85, and 9.23 ms for adults and 18.95, 14.45, 13.45, 10.95, and 12.15 ms for the elderly. Both groups showed common results where the threshold decreased as the frequency increased, with the lowest GDT observed at 4000 Hz; however, the thresholds of the elderly were higher than those of the adults under all conditions. In particular, the threshold variation was found to be larger in the elderly compared to the adults at 500 Hz. When measurements were taken by dividing adults and the elderly into more finely segments (Keith, 2000), the average of all frequencies for young adults was 8 ms, and for the elderly, those in their 60s showed a threshold of 9 ms, which is close to that of adults, but those in their 70s showed a rapidly rising threshold of 22 ms, indicating that auditory temporal resolution ability is closely related to aging.

[0067] Low auditory temporal resolution in elderly individuals with hearing loss results in a decline in speech perception, thus requiring active rehabilitation training. Rehabilitation programs related to central auditory processing and the verification of their effectiveness have been conducted from various angles through numerous studies, and the efficacy of such rehabilitation is being reported. Brain plasticity refers to the continuous neural processing activity that occurs through the interaction between brain neurons by receiving and processing input neural stimuli, leading to reorganization; this process can result in the recovery of damaged functions. In this regard, recent research on auditory temporal resolution has demonstrated that performance improves and is maintained through training. In this study, training was conducted using a method for finding the GDT (Global Diagram Depth). Eighteen individuals from both a group of hearing adults and an elderly individual with hearing were divided into a control group and an experimental group, with nine participants in each group, based on whether they received training. The test used narrowband noise ranging from 917 Hz to 1090 Hz with a stop-band attenuation of 96 dB based on 1000 Hz as the stimulus sound. The total duration of the stimulus sound was 200–300 ms, and the rise and fall times were 20 ms. The stimulus intensity was presented at an intensity 40 dB higher than the average of the subjects' hearing thresholds at 500, 1000, and 2000 Hz. The test method used was the 3 interval three alternative forced-choice (3IAFC) method, which involves selecting a present sound with a gap among three presented sounds. Using adaptive learning (Levitt, 1971), in which the length decreases when there are two correct responses and increases when there is one incorrect response, the gap showing a 70.7% correct response rate on the psychoacoustic functional curve was calculated as the threshold, and the threshold was tested more precisely by measuring the gap in 1 ms increments. The training consisted of a total of 10 tests per session and was conducted in one day.The training was conducted for a total of 10 sessions, with each session lasting one day at intervals of 1 to 2 days. Maintenance evaluations of the training were conducted 24 hours after the last session and one month later. The results of the study showed that initially, the elderly exhibited significantly higher GDT scores than adults; however, after four sessions, their average thresholds approached those of healthy adults, and they maintained the same rate of progress in subsequent training sessions. This suggests that auditory temporal resolution training can prevent the decline in auditory temporal resolution ability in the elderly.

[0068] FIG. 2 is a drawing for explaining the operation of an electronic device according to one embodiment of the present disclosure.

[0069] Referring to FIG. 2, the electronic device (100) can perform a first inspection step of providing a pair of audio data consisting of a first audio data and a second audio data to a user (S210).

[0070] At this time, an electronic device (100) according to one embodiment of the present disclosure may provide a user with the ability to select audio data containing a gap (=gap) from a pair of audio data consisting of a first audio data that matches a first sound, which is an original sound that does not include a gap, and a second audio data that matches a second sound in which at least a portion of the first sound is replaced with a gap (gap).

[0071] Here, the audio data provided to the user by the electronic device (100) as a presentation sound may be composed of a combination of multiple stimulation frequencies and stimulation lengths.

[0072] To this end, the electronic device (100) can generate audio data in which at least one of the stimulation frequency and stimulation length is different.

[0073] Through this, the electronic device (100) can identify the user's interval detection threshold value for each stimulation length by providing audio data with a changed stimulation length at the same stimulation frequency to the user, or the electronic device (100) can identify the user's interval detection threshold value for each stimulation frequency by providing audio data with a changed stimulation frequency at the same stimulation length.

[0074] When first audio data is selected in the first inspection step, the electronic device (100) can perform a second inspection step that increases the gap of the second audio data and provides it to the user (S220).

[0075] Specifically, as a result of providing a pair of audio data consisting of a first audio data and a second audio data through a first inspection step, if a user input selecting the first audio data corresponding to an incorrect answer is received, the electronic device (100) can increase the gap of the second audio data and provide a pair of audio data including the first audio data and the modified second audio data to the user again.

[0076] Alternatively, if the second audio data is selected in the first inspection step, the electronic device (100) may perform a third inspection step that reduces the gap of the second audio data and provides it to the user (S230).

[0077] Specifically, when a user input selecting a second audio data corresponding to the correct answer is received through a first inspection step, the electronic device (100) can reduce the gap between the second audio data and provide the user with a pair of audio data including the first audio data and the modified second audio data.

[0078] And, the electronic device (100) can obtain a gap detection threshold value for the user by repeating the second inspection step and the third inspection step (S240).

[0079] Specifically, when a second inspection step or a third inspection step is performed, the electronic device (100) can repeatedly perform a plurality of inspection steps by repeating the process of selecting the first audio data and the second audio data and repeating the process of changing the gap interval of the second audio data.

[0080] Through a plurality of inspection steps performed, when the gap interval is inverted a certain number of times (e.g., 8 times) according to user input selecting one of the first audio data and the second audio data, a gap detection threshold value is obtained, which is the minimum length gap interval that the user can identify, based on the gap interval of the second audio data at the point in time when it has been inverted a certain number of times.

[0081] Here, the electronic device (100) can obtain the average value of the gap intervals of the second audio data corresponding to each of the most recent preset number of inversion points (e.g., 5 times) based on the inversion point of a certain number of times as the gap detection threshold value. A detailed explanation will be provided later through FIG. 4.

[0082] In particular, an electronic device (100) according to one embodiment of the present disclosure may configure a total of 20 conditions for a presentation sound (= first audio data and second audio data) according to 5 stimulation frequencies (narrowband noise and broadband noise of 500, 1000, 2000, and 4000 Hz) and 4 stimulation lengths (50, 100, 250, and 500 ms), and each condition may include a total of 13 gaps (2 to 40 ms).

[0083] Specifically, the stimulation frequencies consisted of a total of five frequencies: broadband noise (BBN) and narrowband noise (NBN) with center frequencies of 500, 1000, 2000, and 4000 Hz. The criteria for selecting these frequencies prioritized frequencies that are easy to compare with the subjects' hearing and can represent the standards for training, within the frequency ranges that affect speech recognition. Generally, the frequency distribution of Korean phonemes is 233–6767 Hz; based on this, it was decided to select 500, 1000, 2000, and 4000 Hz, which are frequencies used in the subjects' hearing tests, within the range of 200–6000 Hz. In the case of the elderly, hearing loss generally begins at 4000 Hz. In the case of / s,ss / , which has the highest frequency band among Korean phonemes, it shows a range between approximately 4000 and 6000 Hz depending on the vowel it is combined with, and other consonants show a frequency distribution in the 4000 Hz range depending on the vowel combination, so 4000 Hz was selected as the representative frequency.

[0084] Regarding the stimulus length, a total of four lengths were selected: 50, 100, 250, and 500 ms. The lengths were selected based on existing prior studies. A prior study (Schneider, 1999) stated that the minimum stimulus length provided to the user was 2.5 ms and the maximum length was 500 ms, showing threshold fluctuations in the range of 5 to 500 ms. Therefore, 500 ms was selected as the maximum length, and 50 ms was selected as the minimum length considering intercomparability with gap detection tests used in clinical practice. The remaining lengths were selected as 100 ms, which is twice the minimum length, and 250 ms, which is 0.5 times the maximum length.

[0085] At this time, each presented sound was designed to include a gap of 2, 4, 6, 8, 10, 12, 16, 24, 28, 32, 36, and 40 ms, and the test was to start from 20 ms.

[0086] However, the audio data provided to the user is not limited to the conditions described above, and various audio content may be provided to the user based on a diverse range of frequencies, lengths, and spacing intervals.

[0087] At this time, the auditory temporal resolution training of the present disclosure may be performed by using a 2-interval alternative forced choice (2IAFC) method to randomly present a pair of stimulus sounds, one containing a gap and one not containing a gap, with an inter-stimulus interval (ISI) of 1 second. The user is instructed to identify the stimulus sound containing a gap among the first and second stimulus sounds and select the corresponding button. At this time, the training may be performed using adaptive learning in which the length of the gap is reversed based on a correct response of 70.7%.

[0088] For example, starting with a noise presentation condition, if a correct response is shown twice in a row at an interval of 20 ms, the gap length is changed to a level that is one step more difficult, and if a wrong response is shown even once out of two attempts, the gap length is changed to a level that is one step easier.

[0089] In this regard, FIG. 3 is an algorithm for explaining the operation of an electronic device according to one embodiment of the present disclosure performing training according to adaptive learning.

[0090] As illustrated in FIG. 3, the electronic device (100) can provide the first audio data and the second audio data in any order during the first inspection step (S310).

[0091] As described above, when a pair of stimulus sounds containing a gap (second audio data) and stimulus sounds not containing a gap (first audio data) are randomly presented in the first inspection step, and a user input is received in which the user selects the second audio data (S320 - Y), the electronic device (100) may perform the first inspection step again (S330). In this case, the first inspection step may be performed again without changing the gap interval of the audio data.

[0092] Here, if the user selects the first audio data as a result of the first inspection step being performed (S320 - N), or if the first inspection step is performed again and the user selects the first audio data (S340 - N), the electronic device (100) may perform a second inspection step that increases the gap of the second audio data (S350).

[0093] Alternatively, if the first inspection step is performed again and the user input that maintains the selection of the second audio data is received (S340 - Y), the electronic device (100) may perform a third inspection step that reduces the gap of the second audio data (S360).

[0094] When a total of 8 reversals occur as a result of the training according to the above-described process, the training is terminated and the electronic device (100) calculates the average value of the last 5 reversal gap lengths as the gap detection threshold value for the corresponding training session.

[0095] In this regard, FIG. 4 is a diagram illustrating the operation of an electronic device according to one embodiment of the present disclosure to calculate a user's gap detection threshold value as a result of training being performed.

[0096] Referring to FIG. 4, the electronic device (100) can proceed with training by changing the interval, starting with second audio data having a interval of 20 ms.

[0097] Specifically, up to the 7th time, as the correct response selecting the second audio data proceeds, it can be observed that the length of the gap in the second audio data decreases. As disclosed in FIG. 4, the length of the gap decreases only when the correct response proceeds twice.

[0098] On the other hand, from the point when the second audio data is provided to the user 8 times to the point when it is provided 11 times, it can be seen that the length of the gap increases as the user selects the first audio data that does not have a gap (misreaction).

[0099] As the above-described process is repeated, when the user input selecting audio data is reversed a certain number of times (e.g., 8 times), the electronic device (100) can obtain the user's interval detection threshold value by averaging the gap interval of a preset time point (e.g., 5 times) from the gap interval of the second audio data at that time point.

[0100] For example, if a total of 8 reversals occur, the electronic device (100) may stop outputting audio data and calculate the gap detection threshold value of the corresponding training session based on the average value of the last 5 reversal gap lengths.

[0101] Specifically, in FIG. 4, the electronic device (100) can obtain a gap detection threshold value by averaging the reversal gap lengths corresponding to R4 to R8.

[0102] Meanwhile, the electronic device (100) may perform additional screening tests to roughly determine the user's gap detection threshold before or after training, in order to set training parameters of the electronic device (100) or verify the training effect. At this time, the audio data for performing the test may be configured to present 12 different gap intervals (0, 2, 4, 8, 12, 16, 20, 24, 28, 32, 36, 40 ms) as 100 ms wideband noise (BBN) in a total of 33 times, with each interval being presented randomly 3 times according to the 2IAFC method, and the electronic device (100) may calculate the lowest length gap that showed 2 correct responses (66.6%) out of 3 times as the threshold.

[0103] Meanwhile, the electronic device (100) may set a different value for increasing or decreasing the gap interval of the second audio data whenever a test step is performed on the gap interval of the second audio data. In this case, the electronic device (100) may calculate a threshold value for increasing or decreasing the gap interval of the second audio data for the user being tested, based on the gap detection threshold value obtained by multiple users performing the test step.

[0104] Specifically, the electronic device (100) can set the average value of the gap detection threshold values ​​obtained for each of the plurality of users as the expected gap detection threshold value of the user being inspected.

[0105] And, the electronic device (100) can set a threshold value for increasing or decreasing the gap interval whenever a plurality of inspection steps are performed based on a set expected gap detection threshold value.

[0106] For example, the electronic device (100) can set a threshold value to be applied to the next inspection step whenever an inspection step is performed, based on the difference between the gap interval of the second audio data and the expected gap detection threshold value.

[0107] In one embodiment, the electronic device (100) may set the threshold value to a larger value as the difference between the gap interval of the second audio data and the expected gap detection threshold value increases. Conversely, the electronic device (100) may set the threshold value to a smaller value as the difference between the gap interval of the second audio data and the expected gap detection threshold value decreases.

[0108] And, as multiple inspection steps are repeated, the electronic device (100) can change the gap of the second audio data by a threshold value set according to user input selecting either the first audio data or the second audio data.

[0109] By differentially setting threshold values ​​according to the expected interval detection threshold value and the gap interval of the second audio data, the electronic device (100) can set a relatively large difference in difficulty so that the gap interval of the second audio data can reach a value similar to the expected interval detection threshold value as the difference between the expected interval detection threshold value and the gap interval of the second audio data increases.

[0110] Conversely, the electronic device (100) can set the difficulty difference relatively small to accurately obtain the user's interval detection threshold value as the difference between the expected interval detection threshold value and the blank interval of the second audio data becomes smaller.

[0111] At this time, the electronic device (100) can calculate an average value using the interval detection threshold values ​​obtained for all of the multiple users, but it is also possible to group each of the multiple users and calculate an expected interval detection threshold value for each group.

[0112] For example, the electronic device (100) can classify multiple users into multiple groups, such as adults with normal hearing, elderly people with normal hearing, adults with hearing loss, and elderly people with hearing loss, according to user information, and can calculate an expected interval detection threshold value for each group by averaging the interval detection threshold values ​​of the users included in each classified group.

[0113] Additionally, the electronic device (100) can obtain user information of the subject to inspection through user input prior to performing the inspection step, and can identify a group matching the subject to inspection and an expected interval detection threshold value of the group.

[0114] Meanwhile, the electronic device (100) can perform additional auditory time resolution training for each of the user's ears after performing auditory time resolution training for the user.

[0115] In this case, the electronic device (100) can perform auditory time resolution training for each of the user's ears by outputting a pair of audio data that is identical to the first audio data and the second audio data, but in which a voice signal is transmitted to only one of the user's ears.

[0116] And, the electronic device (100) can compare the training results for each of the user's ears and the training results for both of the user's ears, calculate the degree of influence of each of the user's ears related to the training results for both of the user's ears, and provide it to the user.

[0117] Specifically, the electronic device (100) can calculate the degree of influence on each of the user's ears by comparing the difference between the training results for both of the user's ears and the training results for each of the user's left or right ears.

[0118] In one embodiment, it is assumed that a first interval detection threshold value exists as a result of the user's training for both ears, a second interval detection threshold value exists as a result of the user's training for the left ear, and a third interval detection threshold value exists as a result of the user's training for the right ear.

[0119] At this time, if the difference between the first interval detection threshold value and the second interval detection threshold value is smaller than the difference between the first interval detection threshold value and the third interval detection threshold value, the electronic device (100) can calculate the degree of influence on the left ear relatively larger than that on the right ear.

[0120] Alternatively, the electronic device (100) may include at least one artificial intelligence model for calculating the degree of influence of each user's ear.

[0121] The artificial intelligence model of the present disclosure according to this may be based on an algorithm for analyzing the relationship between the auditory temporal resolution training results for each of the user's ears and the auditory temporal resolution training results for both of the user's ears. For example, the artificial intelligence model may be based on algorithms such as a regression analysis model, a recurrent neural network (RNN), and a long short-term memory (LSTM), and may be trained based on the user's training results for both of the user's ears, the training results for each of the user's ears, and information regarding each of the user's ears (e.g., presence or absence of hearing loss, hearing test information, etc.), and when the auditory temporal resolution training results for the user (= first interval detection threshold value to third interval detection threshold value) are input, the degree of influence of each of the user's ears can be calculated.

[0123] 1. Study Participants

[0124] Meanwhile, regarding the application of the electronic device (100) according to the present disclosure, a standard establishment study was conducted on hearing adults and an application study was conducted on the elderly.

[0125] First, the subjects for the study establishing ATR norms for adults consisted of 26 hearing adults (14 males and 12 females) residing in the Seoul, Gyeonggi, and Gangwon regions. The sample size was calculated using the G*Power 3.1.9 sampling program based on Cohen's sampling formula. By inputting an effect size of .25, a significance level of .05, and a power of .80 for a repeated measures one-way ANOVA, the result was 24. To account for potential dropouts, 2 additional subjects—corresponding to 10% of the expected number—were added, resulting in a total of 26 selected. The selection criteria were individuals aged 20 to under 35 years who had no otological diseases, cognitive or neurological problems, and demonstrated normal hearing of 25 dB HL in the octave range of 250–8000 Hz on pure-tone audiometry, with a difference of ±5 dB between the two ears.

[0126] The average age of the 26 hearing adults participating in this study was 24.96±3.0 years. Based on frequencies of 500, 1000, 2000, and 4000 Hz, the pure-tone threshold average (PTA) was 3.17±3.71 dB HL for the right ear and 3.79±4.56 for the left ear, with no significant differences between the ears or genders (p > 0.05).

[0127] Second, ATR threshold values ​​were checked for a total of 13 elderly individuals, including 7 with normal hearing and 6 with hearing loss, using an electronic device (100). The selection criteria for the elderly subjects included individuals aged 65 years or older who had no otological diseases, cognitive problems, or neurological issues, and whose PTA was 25 dB HL or less for elderly with normal hearing and 60 dB HL or less for elderly with hearing loss, based on a frequency of 500–4000 Hz. Additionally, cognitive and neurological tests were conducted on all elderly subjects using the Korean Mini-Mental State Examination (K-MMSE), and subjects with a score of 25 or higher, falling within the normal range, were selected. The average age of the 7 elderly individuals with normal hearing who participated in this study was 68 ± 2.5 years, and their average pure tone hearing threshold was 16.07 ± 7.33 dB HL based on the good ear. The average age of the 6 elderly people with hearing loss was 76.3±7.14 years, and the average pure tone hearing threshold was 40.17±10.71 dB HL according to the good ear standard.

[0128] All participants in the study were informed of the purpose, methods, and procedures of the study and were selected from those who wished to participate voluntarily. This study was conducted with the approval (Approval No. #HIRB-2023-064) of the Institutional Review Board (IRB) of Hallym University regarding all procedures and details of the study.

[0130] 2. Research Procedure

[0131] To screen adult subjects with normal hearing, pure-tone audiometry was performed, followed by ATR threshold testing. Among them, 10 subjects underwent ATR retesting after two weeks, and GIN testing (Musiek et al., 2005) and RGDT testing (Keith et al., 2000) were conducted to evaluate validity. Based on the ATR test results of adults with normal hearing, an application study was conducted on elderly subjects under a 500 ms length condition. For elderly subjects, audiometry and K-MMSE were performed to screen for subjects with normal hearing and those with hearing loss.

[0132] 1) Hearing test and cognitive function test

[0133] Pure-tone audiometry was performed to select study participants. Both tests were measured using an audiometer (GSI 61, Grason Stadler) and headphones. For the elderly, audiometry and speech recognition tests were conducted using a portable audiometer (Audiometer SA 204, Entomed) and headphones. Tests were performed at frequencies of 125, 250, 500, 1000, 2000, 4000, and 8000 Hz. The tests were conducted in a noise environment of 40 dBA or less, and cognitive and neurological assessments were performed using the Korean Mini-Mental State Examination (K-MMSE).

[0134] 2) ATR test

[0135] To test the subject's ATR threshold, an ATR test was conducted in a quiet, soundproof room using the electronic device (100) of the present disclosure. The test was performed using a 2IAFC method to identify a present sound containing a gap among two present sounds, thereby finding a stimulus sound containing a gap among two stimulus sounds. The present sounds were NBN and BBN at 500, 1000, 2000, and 4000 Hz, and the lengths of the stimulus sounds were composed of 50, 100, 250, and 500 ms. The lengths of the gaps were 40, 36, 32, 28, 24, 20, 16, 14, 10, 8, 6, 4, and 2 ms, totaling 13, and the test was started starting from 20 ms. Based on the process described above, adaptive learning was implemented so that if two consecutive correct responses occurred, the test proceeded with a shorter gap, while if even one incorrect response occurred, it proceeded with a longer gap. The test was stopped when eight reversals occurred, where the gap length shifted upward or downward, and the average of the last six gap lengths was calculated as the subject's ATR threshold. Both frequency and length, which were the test elements, were presented in a random order, and the intensity was administered at 50 dB followed by 35 dB.

[0136] 3) Test-retest reliability test

[0137] Meanwhile, in order to determine whether the ATR threshold value of the electronic device (100) according to the present disclosure shows the same threshold even after a period of time, 10 people were randomly selected from 26 hearing adult subjects and participated in the first ATR test, and then a re-test was conducted 14 to 16 days later.

[0138] 4) Validity test

[0139] Ten subjects were randomly selected from among adults with normal hearing and underwent GIN testing (Musiek et al., 2005) and RGDT testing (Keith et al, 2000). The GIN test was performed first, followed by the RGDT test, and thresholds were calculated by randomly presenting RGDT test conditions (500, 1000, 2000, 4000 Hz).

[0140] (1) Gap in noise (GIN) test

[0141] The GIN test was conducted after completing the practice track. It was administered at a presentation intensity of 50 dB HL, identical to the auditory temporal resolution training. During the test, white noise was presented for 6 seconds per trial, and subjects were instructed to press a button whenever a gap of varying lengths—ranging from 0 to 3—was identified. The gap lengths were 2, 3, 4, 5, 6, 8, 10, 12, 15, and 20 ms; three of these were presented randomly, but each length was presented 6 times per track, resulting in a total of 60 gaps. Among the gap lengths, the gap that showed a correct response rate of 4 out of 6 (66.6%) was calculated as the approximate threshold (A.th.). If the correct response rate decreased after the first gap showing 66.6%, the subsequent gap showing 66.6% was calculated as the approximate threshold. The correct response rate for each list was calculated by converting the number of correct responses out of 60 total gap presentations into a percentage, and the average of the four list thresholds was calculated as the final threshold.

[0142] (2) Random gap detection test (RGDT)

[0143] The RGDT was presented at 50 dB HL and administered after sufficient training on the test method using screening and practice tracks. Stimulus sounds were presented at short intervals, and subjects were instructed to press a button to confirm when a short gap included in the middle of the presented stimulus sounds was heard. Stimulus sounds were presented at frequencies of 500, 1000, 2000, and 4000 Hz. Stimulus sounds of 17 ms length were presented as a pair, with a presentation sound containing gaps of 0, 2, 5, 10, 15, 20, 25, 30, and 40 ms in the center, presented once in a random order at 4.5-second intervals.

[0145] 3. Statistical Analysis

[0146] Following the aforementioned testing process, the research results were analyzed. Repeated measures ANOVA was conducted to compare the interactions of ATR thresholds among frequency, stimulus length, and intensity by group, while repeated measures one-way ANOVA was performed to compare ATR thresholds among the elements. Wilcoxn signed-rank tests were conducted to compare differences between elements within each variable. Paired-samples t-tests were performed between the test and retest to verify the reliability of the training program, and Spearman correlation analyses were conducted between the GIN and RGDT results.

[0148] 4. ATR performance in hearing-impaired adults

[0149] Meanwhile, an analysis was performed on 26 healthy adults based on presentation intensity, stimulus length, and stimulus frequency. The results showed that for both conditions of 50 dB and 35 dB, the threshold tended to decrease as the NBN frequency increased from 500 Hz to 4000 Hz. For 50 dB, the threshold decreased with a stimulus length of 50 ms (18.01–4.64 ms), 100 ms (22.92–2.36 ms), 250 ms (25.67–4.23 ms), and 500 ms (26.15–3.5 ms). At 35 dB, the stimulation length decreased from 22.41 to 6.86 ms for 50 ms, 28.16 to 6.4 ms for 100 ms, 26.52 to 6.33 ms for 250 ms, and 30.16 to 7.73 ms for 500 ms.

[0150] Performance across frequency lengths showed that, except for 100 ms under 50 dB and 35 dB intensity conditions, the threshold decreased as the presentation length shortened. For 50 dB, the thresholds for each frequency were 18.01±5.65–26.15±5.41 ms at 500 Hz, 8.61±2.33–10.63±2.56 ms at 1000 Hz, 4.47±1.63–6.07±2.3 ms at 2000 Hz, 4.21±1.79–5.0±2.36 ms at 4000 Hz, and 3.16±0.91–4.21±1.54 ms at BBN. For 35 dB, the duration was 22.41±7.53–30.16±5.85 ms at 500 Hz, 11.81±5–16.61±6.45 ms at 1000 Hz, 5.89±1.88–9.46±2.07 ms at 2000 Hz, 5.15±2.67–7.41±2.97 ms at 4000 Hz, and 6.33±4.37–7.73±5.11 ms for BBN.

[0151] To compare the intensity, length, and frequency of the stimulus, a repeated measures three-way ANOVA was performed. After examining the Greenhouse-Geisser corrected values, a significant interaction was found between frequency * length * intensity [F(5.44, 135.98) = 4.82, p < .001]. Significant interactions were observed between frequency * length [F(5.49, 137.3) = 176.32, p < .001] and between frequency * intensity [F(3.26, 8.16) = 4.51, p < .001], but no significant interaction was found between length * intensity.

[0152] A repeated measures two-way ANOVA was conducted to compare the average performance between frequency and length at intensities of 50 dB and 35 dB, respectively. The results showed that there was a significant interaction between frequency and length for both 50 dB [F(4.93, 165.73) = 397.78, p < .00] and 35 dB [F(5.90, 114.15) = 239.46, p < .001].

[0153] Repeated measures one-way ANOVA was conducted to examine frequency differences across stimulus lengths for each presentation intensity using Greenhouse-Geisser correction values. The results showed significant differences (p < .01) across all lengths under the 50 dB and 35 dB intensity conditions. Paired-sample comparative analysis of frequencies for each presentation intensity (Table IV-1) revealed that for 50 dB, significant differences were observed between frequencies of 500, 1000, and 2000 Hz or higher across all length conditions (p < .01), whereas no significant differences were observed under the 4000 Hz and BBN conditions (p > .05). In other words, regardless of length, the threshold decreased as the frequency increased at frequencies below 2000 Hz (p < .01), and 2000 Hz and 4000 Hz exhibited similarly low gap thresholds across all lengths except for 500 ms. In addition, BBN showed the lowest threshold, but there was no statistically significant difference compared to the 4000 Hz threshold of NBN (p > .05). For 35 dB, regardless of length, there was a significant difference between frequencies of 500 Hz and 1000 Hz or higher (p < .01); however, there was a significant difference between 1000 Hz and 2000 Hz except for 100 and 250 ms (p < .05); there was a significant difference between 2000 Hz and 4000 Hz except for 250 ms (p < .01); and there was no significant difference between 4000 Hz and BBN conditions (p > .05). In other words, as the NBN frequency increased, the significant difference between frequencies decreased. The threshold decreased as the frequency increased from 500 Hz to 4000 Hz regardless of length, and BBN showed the lowest threshold at 100 and 250 ms, while 4000 Hz showed the lowest threshold at 50 and 500 ms, but there was no significant difference (p > .05).

[0154] Repeated measures one-way ANOVA was performed to examine the differences between stimulus lengths at each frequency of presentation intensity using Greenhouse-Geisser corrected values. As a result, at 50 dB, significant differences between lengths were observed at frequencies excluding 4000 Hz (p < .05), and at 35 dB, significant differences between lengths were observed at frequencies excluding BBN (p < .05). As a result of the paired-sample comparative analysis of lengths according to the frequency of each presented intensity, for 50 dB, the significant difference between lengths decreased as the frequency increased in the 500–4000 Hz range; in the case of 500 Hz, significant differences were observed in all length comparisons except for 250 ms and 500 ms (p < .05), whereas for 1000 Hz, 2000 Hz, and BBN, significant differences were observed between 100 ms and other lengths (p < .05), and for 4000 Hz, significant differences were observed only between 100 ms and 250 ms (p < .05). In other words, regardless of frequency, 250 ms and 500 ms did not show a significant difference, while 100 ms showed a significant difference at all frequencies (p < .05). For frequencies below 2000 Hz, the threshold increased with increasing length, but this was not statistically significant for lengths of 250 and 500 ms or longer (p > .05). At 4000 Hz, for BBN, the highest threshold was observed at a length of 100 ms (p < .05), while no significant differences were observed for the remaining lengths (p > .05). For 35 dB, the significant difference between lengths decreased as the frequency increased. From 500 Hz to 4000 Hz, a significant difference was observed between 50 ms and other lengths; at 1000 Hz, a significant difference was observed between 500 ms and other lengths; and for BBN, a significant difference was observed between 250 ms and 500 ms (p < .05). That is, at 500–4000 Hz, there was a significant difference (p < .05) between 50 ms and 500 ms, and the threshold was lower when the length was short.BBN also showed a lower threshold when the length was short, but there was no statistically significant difference (p > .05).

[0155] Repeated measures one-way ANOVA was performed to examine the differences between presentation intensity and length, and between presentation intensity and frequency using Greenhouse-Geisser correction values. The results showed significant differences between intensity and frequency [F(2.53, 24.51) = 510.51, p < .001] and between intensity and length [F(2.45, 14.11) = 3.763, p < .01]. Paired-sample comparative analysis of intensity by frequency and presentation length revealed that, with the exception of 100 ms at 500 Hz and 50 ms at 4000 Hz, all showed significantly lower thresholds at 50 dB (p < .05).

[0156] 1) ATR performance according to frequency

[0157] The ATR threshold of adults with normal hearing decreased as frequency increased at two intensities of 50 dB HL and 35 dB HL, showing the highest performance in BBN. ​​At 50 dB HL, the ATR threshold values ​​for BBN length ranged from 3.16±0.91 to 4.21±1.54 ms for lengths of 50 to 500 ms, which is close to the result of a previous study (Musiek, 2005) of 4.2±0.63 ms. The average performance of ATRs by length for each frequency was 23.19, 9.80, 5.29, and 4.52 ms in the order of 500, 1000, 2000, and 4000 Hz, respectively. In a previous study (Shailer, 1983), the thresholds were approximately 22, 16, 8, 6, 5, and 3 ms in the order of 200, 400, 1000, 2000, 4000, and 6500 Hz, respectively, with the threshold decreasing as the frequency increased in all cases. ATRs showed a difference of more than 10 ms between 500 Hz and 4000 Hz, which was similar to the difference of more than 10 ms between 400 Hz and 4000 Hz observed in a previous study (Shailer, 1983). This appears to be attributed to frequency bandwidth; a narrower bandwidth slows down fluctuation speed, thereby increasing the threshold, and low frequencies require a wider bandwidth compared to high frequencies to detect the same gap length. In this study and a previous study (Shailer, 1983), performance decreased as the stimulus frequency decreased. Both studies showed thresholds of over 20 ms at 400 Hz and 500 Hz, and a difference of 10 ms compared to 1000 Hz, confirming the correlation between bandwidth and low frequency. In the case of auditory filters, the bandwidth is narrower at lower frequencies, requiring a longer stimulus duration compared to other frequencies. However, since ATR and the previous study (Shailer, 1983) provided all stimulus tones uniformly at 500 ms, the results showed a higher threshold at low frequencies, which require a relatively longer presentation length.Although 500 Hz in ATR is a critical frequency for speech recognition, it is a frequency heavily influenced by stimulus factors, and research has confirmed the necessity of training for this frequency. Analysis of the average thresholds at different frequencies showed a significant difference from all other frequencies, suggesting the possibility of training that considers frequency specificity. In particular, 4000 Hz is the primary frequency range where the elderly exhibit hearing loss, highlighting the need to provide ATR training specifically for this demographic. In the case of BBN, because it includes many frequency channels, other frequencies can provide clues to detect a gap even if the gap at 4000 Hz is not recognized, making its use for training specific frequency ranges limited. However, BBN is significant for training threshold detection across overall frequencies. Furthermore, since BBN is utilized in the GIN test—the most widely used clinical test—it offers the advantage of facilitating the immediate and easy establishment of training plans following diagnostic tests.

[0158] 2) ATR performance according to stimulus length

[0159] The threshold difference according to stimulus length showed a significant difference for both intensities at frequencies below 2000 Hz (p < .01). In the correspondence comparison by length, low frequencies were more affected by stimulus length than high frequencies relative to 2000 Hz (p < .05), and lower intensities were more affected by length (p < .05). Both 50 dB and 35 dB intensities showed the greatest significant difference between lengths at 500 Hz, which may be related to acoustic and neurological aspects. Acoustically, this relates to the spectral splatter phenomenon and the aforementioned influence of bandwidth on low frequencies. Spectral splatter refers to the phenomenon where, as the presentation length decreases, energy is dispersed to surrounding frequencies rather than the center frequency, affecting gap detection across other frequencies. Therefore, as the presentation length decreases, the frequency bandwidth of the stimulated signal increases, allowing detection down to a lower threshold. Since the low-frequency range is more affected by the bandwidth, there is a significant difference (p < .05) in the threshold between stimulation lengths.

[0160] The neurological correlation between performance changes and length is neural adaptation. Neural adaptation refers to the phenomenon where neurons become desensitized to specific, repeated stimuli. This adaptation is influenced by the length of the first stimulus and exhibits a negative correlation: as the length increases, the magnitude of the neural response to the stimulus occurring after the gap decreases. This can be understood in connection with the total amount of neural energy. Since neurons have a fixed total amount of energy, increasing the length of the first stimulus consumes more energy, making it impossible to utilize sufficient energy for the second neural stimulus occurring after the gap. Consequently, as the stimulus length shortens, the energy distribution available to the two neural stimuli becomes similar, allowing for better detection of the threshold. Secondly, the reason why low frequencies are more affected by length is also related to adaptation. In the case of high frequencies, the speed of waveform change within the frequency range is rapid; the transformation occurs before adaptation takes place, preventing the neurons from detecting it and thus maintaining a high neural response intensity. However, in the case of low frequencies, the transformation speed is slow, increasing the time required for sound perception. This reinforces adaptation while simultaneously reducing the intensity of the neural stimulus, making the response relatively less distinct. Therefore, adaptation occurs more strongly at low frequencies compared to high frequencies, and since this is influenced by stimulus length, the difference in performance between lengths is greater at low frequencies than at high frequencies.

[0161] When comparing stimulus lengths of 50 ms and 500 ms, significant differences (p < .05) were observed at 500, 1000, and 2000 Hz. This finding is close to that of a previous study (Schneider & Hamstra, 1999) which showed significant differences in thresholds from 500 ms to 50 ms at 2000 Hz. In the previous study (Schneider & Hamstra, 1999), thresholds were measured from 500 ms to 0.5 ms; the threshold actually increased at lengths of 5 ms or less, and at 0.5 ms, the threshold was higher than at 500 ms, confirming that the optimal length range for observing changes in GDT across lengths is 5–500 ms. All four length conditions of ATR fall within this category, and since the threshold decreases with length, length-specific training is required for frequencies of 500, 1000, and 2000 Hz. On the other hand, 4000 Hz showed a significant difference at 50 ms and 100 ms only at 35 dB, while BBN showed a significant difference at 50 ms and 100 ms at 50 dB, and at 500 ms and 250 ms at 35 dB. In a previous study (Formby & Muir, 1989), there was no difference when thresholds were measured with randomly provided lengths in BBN, but the change in threshold according to intensity was not examined. However, in this study, there were differences in performance between lengths depending on intensity in the case of BBN, thus suggesting the need for training for all four lengths.

[0162] 3) ATR performance according to stimulation intensity

[0163] Although this study was conducted on adults with normal hearing, it aimed to indirectly assess the effects on subjects with hearing loss by examining differences in performance according to presentation intensity. Previous studies (Moore, 1993; Shailer, 1983; Plomp, 1963) showed significant differences in auditory temporal resolution performance depending on intensity; while thresholds decreased at higher intensities, no significant difference was observed above a certain intensity level. Each study utilized different stimulus tones and threshold calculation methods; when using pure tones, there was no significant difference in thresholds between intensities starting from 55 dB SPL or higher, while NBN showed no significant difference starting from 40 dB SPL or higher, and BBN from 35 dB SPL or higher.

[0164] However, in this study, the results of tests conducted at 50 dB and 35 dB showed significantly lower thresholds at 50 dB for all tests except for 100 ms at 500 Hz and 50 ms at 4000 Hz (p < .05).

[0165] The two intensity conditions presented in this study fall within an intensity category where there were no significant threshold differences in previous studies, which may be attributed to differences in the type of stimulus sound, threshold calculation methods, and testing methods. However, the fact that performance varies with intensity in this study indicates that when the sound is quiet, auditory temporal resolution and even speech comprehension may be impaired, suggesting that for individuals with hearing loss, auditory temporal resolution may be reduced due to the perceived low intensity of the presented sound.

[0167] 5. Reliability of ATR Training Electronic Devices

[0168] 1) Test-retest reliability

[0169] FIG. 5 is a diagram illustrating the results of a re-examination conducted to verify consistency with the test results after two weeks have passed since an electronic device according to one embodiment of the present disclosure performed a first training session on 10 hearing adults.

[0170] According to Figure 5, the average difference between frequencies by length was 2.48, -1.04, 1.31, 0.64, and 0.6 ms at 50 ms, 0.48, -1.44, -0.24, -0.16, and 0.92 ms at 100 ms, and 4.48, -0.52, 0.96, -0.32, and -0.32 ms at 250 ms, in the order of 500, 1000, 2000, 4000 Hz, and BBN; at 500 ms, it was 3.44, -1.48, 1.32, 0.36, and -0.56 ms. As a result of conducting a paired-sample analysis for test-retest reliability, there were no significant differences in all conditions except for 50 ms at 500 Hz (p > .05). This means that the user's ATR threshold following the auditory temporal resolution training of the present disclosure does not show a difference in performance over time, thus indicating reliability.

[0171] 2) ATR Threshold and Concurrent Validity

[0172] (1) Comparison and Correlation of GIN Threshold and ATR Threshold

[0173] GIN tests were conducted on 10 hearing adults participating in this ATR training. The mean and standard deviation of the GIN test thresholds were 5.50 ± 0.92 ms, and the threshold difference between ATR and BBN ranged from 1.29 to 2.34 ms depending on the length. To compare the GIN test thresholds with the BBN thresholds of ATR at different lengths, a Wilcoxn signed-rank test was performed. The results showed that the BBN threshold of ATR was significantly lower at all lengths except 100 ms (p < .01). Spearman correlation analysis was conducted to check for correlation, and no significant correlation was found between the GIN threshold and the ATR threshold depending on the length.

[0174] (2) Comparison of RGDT and ATR thresholds and correlation

[0175] As a result of administering the RGDT to 10 hearing adults participating in this ATR training, the mean and standard deviation of the RGDT thresholds at 500, 1000, 2000, and 4000 Hz were 5.50±4.6, 5.80±4.46, 5.10±2.92, and 6.70±2.98 ms, respectively. While the RGDT showed a threshold range of 5.10–6.70 ms across all frequencies, the ATR exhibited high thresholds of 8.6 ms or higher at low frequencies below 1000 Hz. A Wilcoxn signed-rank test was conducted to compare the thresholds of the RGDT and ATR at each frequency length; the results showed no significant difference from the RGDT thresholds as the threshold decreased with increasing frequency. At 2000 Hz, no significant difference was observed across all lengths (p > .05), whereas at 500 Hz, significant differences were observed across all lengths. At 1000 Hz, RGDT had a significantly lower threshold than ATR, except for the 50 ms condition (p < .05), while at 2000 Hz and 4000 Hz, no significant difference was observed except for the 500 ms condition at 4000 Hz (p > .05). Spearman correlation analysis was performed on RGDT at 500, 1000, 2000, and 4000 Hz and ATR at 500, 1000, 2000, and 4000 Hz, and no significant correlation was observed under any conditions.

[0177] After auditory temporal resolution training of the electronic device (100) according to the present disclosure was performed, a test-retest was conducted to check the degree of threshold variation over time, and it was proven that reliability was proven as no significant difference was observed in all frequency and length conditions except for 250 ms at 500 Hz. These results were obtained by calculating the threshold once.

[0178] Meanwhile, GIN and RGDT tests were conducted to verify the validity of the auditory temporal resolution training of the electronic device according to the present disclosure. The threshold of the GIN test was 5.5 ms, which is very close to the results of previous domestic studies, with a difference of within 1 ms, compared to the average threshold of 5.03 ms for men and 4.98 ms for women. Prior international studies (Musiek, 2005; Giannela Samelli & Schochat, 2008; John et al., 2012) also showed thresholds of 4.9, 3.9, and 4.7 ms, respectively, which were similar to ATR with a difference of 1 to 1.5 ms, confirming the stability of the test results. Based on this, when compared to the BBN performance of ATR (3.16–4.21 ms), GIN was similar with a threshold 1.3–2.3 ms lower, but no significant correlation was found. Although the thresholds between the tests were similar, there were limitations in the number of subjects to confirm the correlation. However, the fact that the thresholds are similar despite the different testing methods and threshold calculation methods compared to ATR implies that BBN is less affected by testing conditions and methods. At the same time, it suggests the possibility of clinical comparison by indicating that a baseline can be established prior to ATR training and training progress can be evaluated based on the GIN threshold value.

[0179] Unlike ATR, RGDT uses pure tones as stimuli and has short stimulus lengths, which limits the comparison of validity; however, it is significant as the only frequency-specific auditory temporal resolution test tool currently used in clinical practice, so a validity test was conducted. The RGDT results of this study were 5.50±4.62, 5.80±4.46, 5.10±2.92, and 6.70±2.98 ms in order of 500, 1000, 2000, and 4000 Hz, respectively, showing similar thresholds close to 1 ms with the results of a previous study (Heeke, 2018) of 5.62±4.02, 4.95±3.26, 5.24±2.68, and 5.86±2.92 ms. Unlike the GIN test, the threshold difference of RGDT and ATR varied from 0.18 to 21 ms depending on the frequency and length conditions, which can be explained by differences in the stimulation conditions of the test.

[0180] First, when using pure tones, the influence of bandwidth is minimal, so thresholds across frequencies may be similar. However, in the case of ATR's NBN, the bandwidth narrows as the frequency decreases; this reduces frequency cues that influence gap detection, causing the threshold to rise. Consequently, the threshold difference between the two tests increases at lower frequencies. This is supported by results showing that the threshold difference between RGDT and ATR approaches within 2.5 ms at 2000 and 4000 Hz compared to low frequencies. Second, differences in thresholds may occur depending on the stimulus length. When comparing the thresholds of the RGDT (34–74 ms) with those of the ATR (50 ms), which is similar, similar levels were observed with a difference of approximately 2 ms, except at 500 Hz. However, when compared with 500 ms, the difference increased further, demonstrating performance differences caused by variations in stimulus length. For these reasons, no significant correlation could be confirmed between ATR and RGDT, but in the case of 50 ms with similar length conditions, there was a negligible difference of less than 4 ms between RGDT and ATR at 1000, 2000, and 4000 Hz, suggesting that mutual comparisons such as progress evaluation or baseline verification of ATR training are possible.

[0182] 6. ATR performance in the elderly

[0183] The thresholds for each frequency condition were measured in 13 elderly subjects using the MCL with a stimulus length of 500 ms (Table IV-9). The results showed that the thresholds decreased as the frequency increased, with BBN having the lowest value, in the order of 500, 1000, 2000, 4000, and BBN, at 29.85±8.09, 17.05±8.91, 11.6±8.58, 10.76±11.34, and 7.2±4.96 ms, respectively. To examine the differences between frequencies, a repeated measures one-way ANOVA was performed using Greenhouse-Geisser correction values ​​(Table IV-9). The results showed a significant difference between frequencies [F(2.0, 55.313)=0.75, p < .001].

[0184] A mixed two-way ANOVA was conducted to compare performance with adults with normal hearing and based on age and frequency. The results showed that the interaction between age and frequency [F(4, 14.128) = 2.35, p > .05] was not significant, but the main effects of age [F(1, 14.128) = 17.52, p < .001] and frequency [F(4, 14.128) = 162.18, p < .001] were significant.

[0185] FIG. 6 is a diagram for comparing the training results of adults and elderly people for each frequency condition after auditory time resolution training of an electronic device according to one embodiment of the present disclosure has been performed.

[0186] As shown in Figure 6, the performance of adults and the elderly for each frequency condition was compared using the Mann-Whitney comparison test, and the results showed that the elderly exhibited significantly higher thresholds than adults for all frequency conditions (p < .05). Due to limitations in the study period, this study randomly measured auditory temporal resolution performance according to stimulus length and frequency under a 50 dB HL presentation condition, including a 500 ms condition, on 13 elderly subjects, and the number of subjects per condition was limited.

[0187] FIG. 7 is a diagram for comparing the individual ATR thresholds of an elderly person with the average value of a healthy adult as a result of auditory temporal resolution training of an electronic device according to one embodiment of the present disclosure.

[0188] Referring to Figure 7, an examination of the individual score distributions for each condition revealed significant individual differences in performance among the elderly compared to the performance of adults with normal hearing. The proportion of elderly subjects exhibiting performance levels of -2 SD or lower than the average adult threshold was 25%, 36%, 42%, 33%, and 46% for 500, 1000, 2000, 4000 Hz, and BBN, respectively, showing a proportion of more than one-third for all frequencies except 500 Hz. Among these, the proportion of elderly individuals with hearing loss was 50%, 25%, 80%, 100%, and 67% for 500, 1000, 2000, 4000 Hz, and BBN, respectively; this figure increased with higher frequencies. Although the figure for 500 Hz was 50%, it represents one in two individuals, making it difficult to compare based solely on ratios.

[0190] In conclusion, after the auditory temporal resolution training of the electronic device (100) according to the present disclosure was performed, the ATR performance of the elderly and adults was compared, and the difference in threshold between frequencies was significant at frequencies other than 500 Hz. Similar to adults, the threshold decreased as the frequency increased from 500 to 4000 Hz, and the BBN threshold was the lowest. In all frequency conditions, the thresholds of the elderly were higher than those of adults, which was a similar result to that of prior studies (Snell 1997; Schneider, 1998; Strouse, 1998; Snell & Frisina, 2000; Harris, 2010; Park & ​​Lee, 2016). When examining the ATR performance of the elderly in previous studies and this study by frequency conditions, for BBN, a domestic study (Park & ​​Lee, 2016) showed 6.6 ms for elderly with normal hearing and 8.8 ms for elderly with hearing loss, while a study conducted with a wide bandwidth stimulus sound of 5 kHz or less (Harris, 2010) showed 5.16 ms; both studies showed values ​​close to ATR with a difference of 2 ms. In other frequencies, at 500 Hz (Tyler, 1982), the score was 22.9 ms, which was 6 ms lower than ATR, and at 1000 Hz (Strouse, 1998; Snell, 2000; Lister, 2005), the scores were 7–10 and 3.4–10 ms, which were 7–10 ms lower than ATR. At 2000 Hz (Schneider, 1999; Schneider, 2000), the values ​​were 3–6 and 4–8 ms, which were 1–7 ms lower than the ATR, while at 4000 Hz (Tyler, 1982), the values ​​were 11.6 ms, which was 1 ms higher than the ATR. Previous research results varied by frequency, but the difference from the ATR was within 1–10 ms.Compared to 500–2000 Hz, the difference from ATR for BBN and 4000 Hz was close at 1–2 ms. This implies that, just as the difference in performance between lengths was minimal in 4000 Hz and BBN in this study, high frequencies are less affected by stimulus factors, meaning that conducting research using different methods does not significantly impact performance. However, it can be seen that low frequencies showed differences in performance because factors such as bandwidth, length, and type of stimulus varied across studies. Consequently, it is expected that the decline in temporal resolution ability in the elderly will be further exacerbated in everyday life environments where various stimulus sounds are presented.

[0191] An interesting finding in the comparison with a previous study (Tyler, 1982) is that although the hearing of elderly subjects with hearing loss at 4000 Hz was 73.6 dB—approximately 30 dB higher than in this study—there was little difference in performance. This indicates that when comparing the subjects' hearing with GDT thresholds, those with hearing of 60 dB or higher generally showed thresholds higher than the average adult range; however, at 500 Hz, there were cases where the GDT threshold was high even with good hearing, suggesting that the association between hearing loss and auditory temporal resolution is not clear. In this study as well, there were elderly subjects with normal hearing who showed a decline in performance of less than -2 SD compared to the average of normal hearing adults, regardless of their hearing level; however, the proportion of elderly subjects with hearing loss increased as the frequency increased, showing ratios of 50%, 25%, 80%, 100%, and 67% at 500 Hz, 1000 Hz, 2000 Hz, 4000 Hz, and BBN, respectively, indicating an association with hearing loss. Another study (Lister, 2005) also found differences between elderly individuals with hearing loss and those with normal hearing. Based on this, it was confirmed that while there may be differences in GDT performance due to aging, the factors attributable to hearing loss are not certain, and there is a possibility of a decline in auditory temporal resolution due to aging regardless of hearing loss. Therefore, the need for appropriate rehabilitation based on auditory temporal resolution screening results rather than hearing test results is raised. Furthermore, since low frequencies below 2000 Hz show high variability in GDT depending on stimulus elements, auditory temporal resolution ability may fluctuate depending on the various stimuli heard in daily life; conversely, perception ability for low-frequency sounds may also fluctuate, suggesting the need for individual training covering frequencies from high to low.

[0192] Meanwhile, the elderly required more time than adults due to a decline in responsiveness and comprehension regarding the auditory temporal resolution training provided by the electronic device (100). Therefore, about 4 to 5 audio data points were suitable for testing, but it was difficult to train all 20 stimulus conditions. Therefore, in order to pre-evaluate the performance of the elderly, it is necessary to select and measure the frequency and length conditions that best represent the threshold performance of the elderly, and individual training tailored to this is required. Although the threshold of 500 Hz in frequency conditions is higher than that of other frequencies, it is easy to train by comparing performance under conditions of the same length and bandwidth. However, it was confirmed that there are difficulties in comparing performance using other tests used in clinical practice as a scale. In the case of other frequencies, the threshold difference from other tests is not significant, so baseline information for ATR training can be provided, but direct application is not easy in the case of 500 Hz. Therefore, it was confirmed that there is a need for further development to identify bandwidth and stimulus length levels that show a threshold similar to other frequencies and use them as stimulus sounds for additional training or screening tests, making comparison and interaction with other tests easier. The necessity of screening tests can be seen in the results showing a decline in temporal resolution ability in hearing-free older adults; measuring performance through screening tests for auditory temporal resolution rather than hearing ability will lay an accurate foundation for determining the necessity and direction of future training.

[0194] 7. Significance as a Training Program

[0195] While there are few prior studies on training adults or the elderly using auditory temporal resolution training programs, the study by Kishon-Rabin (2013) applied an adaptive training method, such as ATR, to hearing adults and the elderly and demonstrated that the training was effective for rehabilitation. In this study, only 1000 Hz NBN stimuli were used, which could not provide training for vulnerable frequencies in subjects with high-frequency hearing loss, whereas ATR can provide stimuli of various frequencies. Regarding gap length, Kishon's study used 1 to 20 ms, which is feasible for hearing-free adults but may present limitations for those with hearing loss. However, ATR provides various gaps up to 40 ms, allowing for training even for subjects with hearing loss. In terms of testing methods, Kishon used the 3IAFC method, employing a more complex testing technique than ATR by measuring gaps more precisely in 1 ms increments. Nevertheless, the fact that the subjects showed progress in training indicates that training effects can be expected even through simpler methods. While ATR was conducted once for norm verification, Kishon structured the training to include 10 threshold tests per session and evaluated progress and maintenance after continuous training at intervals of 1-2 days. As a result of the training, the elderly demonstrated the effectiveness of the training by showing a gradual decrease in thresholds up to the 10th day of training; although their thresholds were about 1-2 ms higher than those of hearing adults, they showed similar thresholds. Performance was maintained with a difference of about 1-2 ms even after one month, and this was consistent with the adult subjects. Therefore, this study suggests the possibility that repetitive training using ATR can improve auditory temporal resolution ability and implies that training effects can be expected not only for hearing-impaired elderly but also for hearing-deaf elderly, depending on the diversity of ATR testing conditions.

[0197] Meanwhile, the electronic device (100) described above can be connected to at least one user terminal (200) and implemented as an auditory time resolution training system.

[0198] In this regard, FIG. 8 is a drawing for illustrating an auditory time resolution training system according to one embodiment of the present disclosure.

[0199] Referring to FIG. 8, the electronic device (100) can communicate with a user terminal (200).

[0200] In this case, instead of outputting training content through the audio output unit (120) and the display (130), the electronic device (100) can communicate with the user terminal (200) through the communication unit as described above to provide training content to the user terminal (200) so that the user can perform training.

[0201] At this time, as training content is provided to the user terminal (200), user input regarding the training process described above can be provided from the user terminal (200) to the electronic device (100).

[0202] A user terminal (200) can communicate with an electronic device (100) via wired or wireless connection, and multiple terminal devices can be simultaneously connected to the electronic device (100). Additionally, the user terminal (200) can be connected to the electronic device (100) through a dedicated program, application, or web browser.

[0203] At this time, the user terminal (200) may be various types of terminal devices capable of connecting to wired / wireless networks, such as smartphones, tablet PCs, desktops, laptops, and wearable devices, but is not limited thereto.

[0204] Meanwhile, the various embodiments described above may be implemented by combining two or more embodiments, provided that they do not conflict or contradict each other.

[0205] Meanwhile, the various embodiments described above may be implemented in a recording medium readable by a computer or a similar device using software, hardware, or a combination thereof.

[0206] According to hardware implementation, the embodiments described in this disclosure may be implemented using at least one of ASICs (Application Specific Integrated Circuits), DSPs (digital signal processors), DSPDs (digital signal processing devices), PLDs (programmable logic devices), FPGAs (field programmable gate arrays), processors, controllers, microcontrollers, microprocessors, and other electrical units for performing functions.

[0207] In some cases, the embodiments described herein may be implemented as the processor itself. In a software implementation, embodiments such as the procedures and functions described herein may be implemented as separate software modules. Each of the aforementioned software modules may perform one or more functions and operations described herein.

[0208] Meanwhile, computer instructions or computer programs for performing processing operations in electronic devices, such as robots and servers, according to the various embodiments of the present disclosure described above, may be stored on a non-transitory computer-readable medium. When such computer instructions or computer programs stored on the non-transitory computer-readable medium are executed by the processor of a specific device, the specific device described above performs the processing operations in the electronic device according to the various embodiments described above.

[0209] A non-transient computer-readable medium refers to a medium that stores data semi-permanently and can be read by a device, unlike media that store data for a short period of time such as registers, caches, and memory. Specific examples of non-transient computer-readable media include CDs, DVDs, hard disks, Blu-ray discs, USBs, memory cards, and ROMs.

[0210] Although preferred embodiments of the present disclosure have been illustrated and described above, the present disclosure is not limited to the specific embodiments described above. It is understood that various modifications can be made by those skilled in the art without departing from the essence of the present disclosure as claimed in the claims, and such modifications should not be understood individually from the technical spirit or perspective of the present disclosure. Explanation of the symbols

[0211] 100: Electronic device 110: Memory 120: Audio output section 130: Display 140: User Input Section 150: Processor 200: User terminal

Claims

Claim 1 A method of operation of an electronic device for performing auditory temporal resolution training for the rehabilitation of a user’s speech recognition ability, comprising: a first test step of providing a user with a pair of audio data, the pair of audio data, the first audio data matching a first sound and the second audio data matching a second sound in which at least a portion of the first sound is replaced with a blank space; a second test step of, when user input corresponding to an incorrect answer of selecting the first audio data in the first test step is received, increasing the blank space of the second audio data and providing the user with a pair of audio data including the first audio data and the second audio data with the increased blank space; and a third test step of, when user input corresponding to a correct answer of selecting the second audio data in the first test step is received, decreasing the blank space of the second audio data and providing the user with a pair of audio data including the first audio data and the second audio data with the decreased blank space. Claim 2 The method of operation of the electronic device according to claim 1 comprises: a step of performing a plurality of inspection steps while repeating a process of changing the gap interval of the second audio data according to a user input selecting one of the first audio data and the second audio data, starting with the second inspection step; and a step of obtaining a gap detection threshold value, which is a minimum length gap interval identifiable by the user, based on the gap interval of the second audio data at the point in time when the user input selecting one of the first audio data and the second audio data is inverted a certain number of times through the plurality of inspection steps. Claim 3 A method of operation of an electronic device according to claim 2, wherein the step of obtaining the interval detection threshold value is to obtain the user's interval detection threshold value according to the average value of the gap interval of the second audio data identified for each of the most recent preset number of inversion points based on the point in time that has been inverted by a predetermined number of times. Claim 4 The method of operation of the electronic device according to claim 1 comprises: a step of performing a plurality of inspection steps while repeating a process of changing the gap interval of the second audio data according to a user input selecting one of the first audio data and the second audio data, starting with the third inspection step; and a step of obtaining a gap detection threshold value, which is a minimum length gap interval identifiable by the user, based on the gap interval of the second audio data at the point in time when the user input selecting one of the first audio data and the second audio data is inverted a certain number of times through the plurality of inspection steps. Claim 5 In claim 1, the first inspection step receives a user input selecting one of the first audio data and the second audio data by providing the first audio data and the second audio data in any order, and the method of operation of the electronic device performs the second inspection step of increasing the gap of the second audio data when the user input of the user corresponding to an incorrect answer selecting the first audio data is received in the first inspection step, and when the user input of the user corresponding to a correct answer selecting the second audio data is received in the first inspection step, the first inspection step is performed again, and when the user input of the user corresponding to an incorrect answer selecting the first audio data is received in the first inspection step performed again, the second inspection step of increasing the gap of the second audio data is performed, and when the user input of the user corresponding to a correct answer selecting the second audio data is received in the first inspection step performed again, the third inspection step of decreasing the gap of the second audio data is performed. Claim 6 The method of operation of the electronic device according to claim 1 comprises the step of generating a plurality of first audio data in which at least one of the stimulation frequency and the stimulation length is different, based on a plurality of stimulation frequencies and a plurality of stimulation lengths; and the method of operation of the electronic device further comprises performing at least one of the first inspection step, the second inspection step, and the third inspection step based on each of the plurality of first audio data. Claim 7 In claim 1, the second inspection step increases the gap interval of the second audio data by a first threshold value, and the third inspection step decreases the gap interval of the second audio data by a second threshold value; the method of operation of the electronic device comprises: a step of setting an expected gap detection threshold value of the user based on a gap detection threshold value obtained for each of the plurality of users; and a step of setting the first threshold value and the second threshold value whenever a plurality of inspection steps are performed based on the set expected gap detection threshold value; wherein the step of setting the first threshold value and the second threshold value sets the first threshold value and the second threshold value to a larger value as the difference between the gap interval of the second audio data and the expected gap detection threshold value is larger, and sets the first threshold value and the second threshold value to a smaller value as the difference between the gap interval of the second audio data and the expected gap detection threshold value is smaller. Claim 8 An electronic device for performing auditory temporal resolution training for the rehabilitation of a user's speech recognition ability, comprising: a memory storing a plurality of sounds for auditory temporal resolution training; a processor that performs a first test step of providing a user with a pair of audio data, the first audio data matching a first sound and a second audio data matching a second sound in which at least a portion of the first sound is replaced with a blank; when user input corresponding to an incorrect answer of selecting the first audio data in the first test step is received, increases the blank interval of the second audio data and provides a pair of audio data including the first audio data and the second audio data with the increased blank interval to the user; and when user input corresponding to a correct answer of selecting the second audio data in the first test step is received, decreases the blank interval of the second audio data and provides a pair of audio data including the first audio data and the second audio data with the decreased blank interval to the user. Claim 9 An electronic device for an auditory temporal resolution training system for the rehabilitation of a user's speech recognition ability, comprising: a first test step of providing a pair of audio data to a user terminal, the pair of audio data comprising a first audio data matched to a first sound and a second audio data matched to a second sound in which at least a portion of the first sound is replaced with a blank space; a second test step of increasing the blank space of the second audio data and providing a pair of audio data including the first audio data and the second audio data with the increased blank space of the second audio data to the user terminal when a user input corresponding to a correct answer selecting the second audio data is received from the user terminal in the first test step; and a third test step of decreasing the blank space of the second audio data and providing a pair of audio data including the first audio data and the second audio data with the decreased blank space of the second audio data to the user terminal when a user input corresponding to a correct answer selecting the second audio data is received from the user terminal in the first test step. A system comprising: a user terminal that receives a pair of audio data from the electronic device and provides a user input to the electronic device for selecting audio data. Claim 10 A non-transient computer-readable medium storing at least one instruction that is executed by a processor of an electronic device to cause the electronic device to perform the method of operation of claim 1.

Images