New hearing test system

Abstract

[Solution] The present disclosure discloses a software unit intended for auditory evaluation, the software unit being arranged to perform a method comprising: performing a reference frequency audiometry test to establish a reference frequency for each ear of a subject; and performing an equal-loudness audiometry test for each ear, the step of performing a reference frequency audiometry test comprising: starting with a first frequency in a given frequency range; providing the subject with a sound at a set volume which is a minimum test volume selected from the range of -10 to 50 Phon to check for the presence or absence of an auditory response; if the subject shows an auditory response at a minimum test volume selected from the range of -10 to 50 Phon, changing the volume by the first frequency and then selecting the first frequency as the reference frequency; otherwise, selecting another frequency in a given frequency range and repeating the iterative procedure with new frequencies in a given frequency range until the software unit detects a frequency X which is the frequency at which the subject showed an auditory response at a minimum test volume selected from the range of -10 to 50 Phon or at which the subject showed a minimum volume auditory response to the set frequency being tested, and then selecting frequency X as the reference frequency.

Description

Background Art

[0001] Generally used ISO standard audiometry typically involves measuring an individual's auditory threshold at a limited number of frequencies, usually around 6 to 10. The results of the test are recorded on an audiogram. This test is useful for detecting an individual's hearing loss by revealing an increase in the auditory threshold, but it does not show any information regarding hearing loss beyond the auditory threshold. Any hearing restoration applied based on the test has to rely on the assumption that hearing loss is linear at high sound pressure levels and approximately equal to the increased threshold. It has been shown from the newly provided equal-loudness audiometry that this is clearly not the case. The hearing ability of individuals suffering from hearing loss varies significantly at sound pressure levels beyond the auditory threshold.

Brief Description of the Drawings

[0002]

[0003] Figure 1 shows equiloudness audiometric test data at 10 Phon, 20 Phon, 30 Phon, and 40 Phon levels for an individual with particularly pronounced high-frequency hearing loss. The bottom trace represents the smallest 10 Phon level at 800 Hz, which is slightly above or equal to the subject's hearing threshold. On the right, the traces show a significant increase from a state where the hearing threshold is slightly above 1 kHz. On the left side of the figure, at low frequencies, the traces are spaced at 10 Phon intervals as expected. At high frequencies, starting at approximately 2 kHz, this interval is significantly smaller than 10 Phon. At 300 kHz, the 10-40 Phon traces are spaced about 40 dB apart, which corresponds to an increase in noise level. A slight increase in the 10 Phon traces between 100 Hz and 1 kHz causes the interval between the 10 Phon and 20 Phon traces to be narrower than the intervals between other traces. This is due to a slight increase in the hearing threshold that affects the smallest 10 Phon level. Above 20 Phon, the increase disappears, and the interval becomes exactly 10 Phon, with only slight local deviations. At high frequencies above 2 kHz, where there is a significant increase in the auditory threshold, the 10-40 Phon traces are separated by only a few Phons. This makes it clear that, for example at 30 kHz, the sound pressure level only needs to increase by a few Phons for the subject to perceive a 40 Phon increase. It is clear that once the sound pressure level exceeds the auditory threshold, the brain rapidly begins to compensate for the elevated auditory threshold, and the ear / brain no longer functions as a linear device, clearly refuting the conventionally accepted assumption that hearing loss is linear.

[0004] This hearing loss behavior poses a significant problem when compensated for by current standard hearing methods that rely on the assumption of linear auditory behavior, given that the elevation of the hearing threshold is amplified. Using standard compensation methods that statically amplify the measured elevated hearing threshold by approximately one-half to two-thirds, sounds are reliably under-amplified at low noise levels and frequencies above approximately 2 kHz, and over-amplified at higher levels. This is evident from the fact that subjects whose hearing was tested by equal-loudness hearing tests, as shown in Figure 1, do not use professionally fitted hearing aids because they do not perceive any benefit from them.

[0005] Typically, dynamic range compression at frequencies with elevated auditory thresholds is smaller than that shown in Figure 1, but all subjects measured exhibited similar dynamic range compression at frequencies with elevated auditory thresholds. The collected measurement data clearly shows that the brain automatically compensates for elevated auditory thresholds at high sound pressure levels when sound exceeds the auditory threshold and is therefore audible. This brain compensation is noticeable in all cases, though it can be very dramatic or less pronounced, as shown in Figure 1.

[0006] To not only assess the presence or absence of hearing loss, but also to obtain data that enables appropriate hearing recovery, a device is needed that includes software to facilitate novel testing methods. For this purpose, the novel Equal Loudness Auditory Test (ELHT) employs several additional testing steps compared to the standard ISO auditory threshold test. The following description applies equally to the device, software, and method for performing the ELHT.

[0007] Apart from the most important aspect of enabling proper hearing recovery, equal-loudness hearing tests also efficiently utilize the available dynamic range. ISO standard hearing tests and compensation based on static amplification consume a large portion of the available dynamic range without any benefit. In fact, the static amplification method exposes the user to unnecessarily and disadvantageously high sound pressure levels, thus increasing the risk of further hearing loss. For an individual with hearing loss, as shown in Figure 1, where the hearing threshold elevation on a standard test is approximately 70 dB at 3 kHz, the usual method is to amplify the sound at that frequency by about half to two-thirds of the measured elevation. In this case, it is about 35 dB to 45 dB, and for example, 40 dB is used.

[0008] As shown by the equal-loudness hearing test in Figure 1, a 40 dB amplification is insufficient at volumes below 30 phons, correct at 30 phons, and excessive at higher volumes. At approximately 70 dB, as is clear from the equal-loudness hearing test data in Figure 1, no further amplification is necessary, and a static 40 dB amplification is extremely excessive. Therefore, it is no surprise that the hearing measurements of individuals shown in Figure 1 do not improve with conventional methods.

[0009] In equal-loudness hearing tests with dynamic amplification that changes according to volume, no additional amplification is applied to the user at volumes above 70 phons. Dynamic hearing recovery devices eliminate the risk of further hearing loss caused by unintentional and unnecessary high volumes resulting from hearing loss compensation.

[0010] Furthermore, regarding the utilization of available dynamic range, for example, a personal music application on a mobile device equipped with Bluetooth-connected earphones or headphones already has a fairly limited dynamic range, restricted by both the headphone hardware and the Bluetooth connection. The digital lower limit is 16-bit resolution, i.e., a dynamic range of 96dB, and in reality, it is even lower, coupled with hardware limitations. If 40dB of that dynamic range is already used in static amplification, then at most 56dB remains. Assuming the maximum output of Bluetooth headphone software is 100dB, it cannot reproduce any sound below 44dB, which is the limit of the digital resolution floor, not the noise floor. Sounds fainter than 44dB cannot be reproduced, meaning that many faint sounds become completely inaudible, and the system's purpose, namely hearing recovery to improve clarity, cannot be achieved. This is also true for hearing aids, which typically operate with a more limited bit depth. [Overview of the project]

[0011] The basis of ELHT is equal loudness. The ISO 226 standard provides equal loudness information at different volume and frequency levels, and ELHT uses this information to assess hearing loss. Gain table data is output from ELHT to a real-time signal processing unit. The real-time signal processing unit includes a compensating peak filter at each frequency where amplification is needed to restore hearing. The compensating peak filter dynamically changes the boost at the filter frequency in response to the instantaneous sound pressure at that frequency. The gain table data output from ELHT includes compensating gain information for all practical sound pressure levels at each frequency.

[0012] The focus of the method described herein is to establish at least one reference frequency for each ear. In its most common form, the Disclosure relates to a software unit intended for auditory evaluation, the software unit being configured to perform a method comprising: performing a reference frequency audiometry test to establish a reference frequency for each ear of a subject; and performing an equal-loudness audiometry test for each ear, the method comprising: starting with a first frequency in a given frequency range; providing the subject with a sound at a set volume which is a minimum test volume selected from the range of -10 to 50 Phon to check for the presence or absence of an auditory response; if the subject shows an auditory response at the minimum test volume selected from the range of -10 to 50 Phon, changing the volume by the first frequency and then selecting the first frequency as the reference frequency; otherwise, selecting another frequency in a given frequency range and repeating the iterative procedure with new frequencies in a given frequency range until the software unit detects a frequency X which is the frequency at which the subject showed an auditory response at the minimum test volume selected from the range of -10 to 50 Phon or at which the subject showed a minimum volume auditory response to the set frequency being tested, and then selecting frequency X as the reference frequency.

[0013] While this disclosure focuses on software units intended for auditory assessment, corrections for one or more hearing impairments may be provided.

[0014] The first ELHT step aims to establish reference frequencies for equal loudness evaluation. One reference frequency is required for each ear. The frequencies may be the same for both ears, but in some cases, it is beneficial to use two different frequencies. The perceived loudness of the reference frequency at different volume levels is used to adjust for the perceived loudness at all other frequencies during the test.

[0015] The second ELHT process adjusts the perceived sound of the reference frequency between the two ears. Since all measurements are adjusted to the perceived volume of the reference frequency, the levels in both ears must be the same at all sound pressure levels. If they are different, the entire spatial perception of sound in the whole space becomes distorted, resulting in a decrease in spatial auditory ability.

[0016] This second step is not essential, but is preferably performed in accordance with the present disclosure.

[0017] The third ELHT step is equal loudness testing. The perceived loudness of a reference frequency, one for each ear, is compared to a set of test frequencies suitably distributed across a frequency range of multiple appropriate sound pressure levels above the auditory threshold.

[0018] Furthermore, steps 4, 5, and 6 described below are optional according to this disclosure. The focus of the method according to this disclosure is to establish at least one reference frequency for each ear.

[0019] The fourth step converts the equal-loudness data measured by Phon into sound pressure level (SPL). The output from the third ELHT step is Phon. SPL data is necessary for efficient signal processing in hearing recovery devices such as earphones and hearing aids. SPL is a physical unit usually measured in dBSPL, a logarithmic unit that uses a sound pressure of 20 micropascals as the reference level, while Phon is a logarithmic perceptual unit. The perceptual level measured by Phon is based on the loudness of sounds that humans perceive as being close.

[0020] The fifth ELHT data post-processing step first adds missing auditory correction data. The measurements performed in step 3 typically cover only a limited dynamic range of 10 to 60 phons. For proper auditory correction, data in the entire relevant dynamic range from 0 dBSPL to 110 dBSPL is required. The correction gain data for the missing portion of the dynamic range is calculated by polynomial extrapolation of the measurement range that can be obtained at all frequencies. The next step involves selecting suitable correction frequencies to which the correction gain should be applied. Frequencies to which correction gain is not needed depending on the measurement are discarded. Frequencies to which hearing loss is severe and correction is not feasible are also discarded. This typically occurs at the highest frequencies of 9600 Hz and 12800 Hz, where the frequency cannot be heard due to age-related hearing loss, and therefore correction is meaningless and, in fact, could be harmful if excessive gain is applied. Finally, the remaining correction gain data is analyzed to find suitable peaks for applying correction in the hearing loss data. Typically, a single peak frequency can be covered by the bandwidth of the corrective peak filter, thus eliminating the need to boost the adjacent frequencies closest to the center frequency. Once the appropriate frequency is selected, two final steps are taken, including limiting the maximum gain to avoid acoustic feedback in the hearing correction device, and limiting the maximum sound output at the gain, and therefore at the maximum sound pressure.

[0021] The sixth ELHT process is tuned for collective amplification. Amplifications from adjacent frequencies are interacted and added together to generate additional gain. The bandwidth of the correction filter, and therefore the bandwidth of the amplification, must be wide to obtain a good time-domain response while limiting energy dispersion, i.e., ringing. Therefore, a wide bandwidth causes overlap between adjacent tuned frequency bands. Optimization must be performed to correct the collective gain to equal the measured required gain. This optimization is called gain-sail optimization because the applied gain appears like a sail on a three-dimensional plot with frequency, volume, and amplification as axes, as shown in Figure 6. [Modes for carrying out the invention]

[0022] The ELHT system differs significantly from current ISO standard hearing threshold tests, which only provide information about the auditory threshold, by employing a novel hearing test that assesses hearing loss over a wide dynamic range. ELHT is based on the perceived equal loudness of a reference tone to a test tone. The test volume is adjusted until the subject perceives the sound as being the same loudness as the reference tone. Typically, the reference tone and test tone are played alternately at an alternative frequency of approximately 0.5 Hz. It is evident that the alternative frequency can be changed to either a higher or lower frequency. A feasible range is 0.25 Hz to 1 Hz. The timbres played alternately for comparison may be automatically or manually controlled. The equal loudness levels used are based on ISO 226 standard levels at different frequencies and sound pressures. The test tone and reference tone may be pure sine waves, or other types of sounds whose bandwidth is limited to the desired test frequency band. Complex test signals with a wider frequency spectrum than pure sine waves provide more information about hearing ability in the test frequency band and are particularly useful in tests with a limited test frequency band.

[0023] Common auditory threshold tests use pure sine waves, which poses a significant problem for individuals suffering from tinnitus. With tinnitus, it becomes very difficult to distinguish between the perceived tinnitus sound and the test sounds generated to assess auditory ability, leading to confusion and inaccurate test results.

[0024] To improve accuracy and ease of use, vibrations are used in ELHT, which are useful as an alternative to pure sine waves. Of course, many other types of multi-frequency timbres and noises may be used, but vibrations have several beneficial characteristics. First, vibrations are not confused with tinnitus. Vibrations are sounds that are significantly different from pure sine waves perceived by people experiencing tinnitus. Also, vibrations have been scientifically evaluated to yield very similar auditory threshold test results compared to pure sine waves. Vibrations have a bandwidth that extends just above and just below the test frequency, and are therefore somewhat less sensitive to local dips and peaks in the frequency response generated by the test equipment and the user's ear and auditory canal. This is particularly suitable at higher frequencies where dips and peaks are more likely to occur. Finally, vibrations are easier to judge in terms of loudness than pure sine waves, and equal loudness comparisons are easier, especially in equal loudness testing. The appropriate modulation range for vibrations is 1% to 10%, and 5% is typically used in ELHT. The standard ELHT modulation frequency is 15 Hz, and the range of 5 Hz to 30 Hz is useful.

[0025] The equal-loudness audiometric test shown in Figure 1 uses 81 frequencies in the range of 100 Hz to 10 Hz. Such a large number of correction filter frequencies are not necessary in an equal-loudness audiometric test. The number of correction filter frequencies can be any number from 1 to the presented 81 or more. Preferably, the correction filter frequencies are 100 Hz, 200 Hz, 400 Hz, 800 Hz, 1200 Hz, 1600 Hz, 2400 Hz, 3200 Hz, 4800 Hz, 6800 Hz, 9600 Hz, and 12800 Hz. These frequencies are evenly distributed across the frequency range relevant to hearing recovery. They are more concentrated at higher frequencies where higher resolution is required for optimal correction.

[0026] The ELHT is performed at a sound pressure level interval of 10 Phon, which is just above the hearing threshold of the subject at the reference frequency where the minimum inspection level is selected. Any other sound pressure level granularity may be used, but if the interval is narrower than 10 Phon, there is no advantage in terms of accuracy and the burden on the subject will increase significantly. In the ELHT test, the interval may be widened in order to more quickly find a specific hearing threshold limit. In such a case, the oversight right is 20 Phon. Also, the granularity may be arbitrary, but an interval of 10 Phon is a good compromise in terms of the balance of accuracy, ease of use, and speed.

[0027] Figure 2 shows the graphical user interface (GUI) of the ELHT software application. The figure is the GUI of step 1 with functions for checking and setting the correction table for the DSP hearing correction device.

[0028] If there is no hearing impairment or it is low for the subject at the selected reference frequency, the minimum ELHT level is usually set to k10 Phon. If the subject has hearing impairment at the selected reference frequency, the minimum level may be increased. The minimum level must be just above the hearing threshold of the subject at the reference frequency, that is, above 0 - 10 Phon. To evaluate the listening ability, at least one higher sound pressure level is required. The more sound pressure test levels there are, the better the evaluation of the subject's listening ability will be. Usually, four test levels with a 10 Phon interval of sound pressure can give acceptable results. If the level is more than that, the burden on the subject during the test will clearly increase significantly, but the listening ability can be better understood over a larger dynamic range. It can be well understood.

[0029] - Step 1. Selection of reference frequency The purpose of the reference selection step is to find a small number of reference frequencies for each ear that have no significant hearing impairment and the listening ability is as normal as possible. Each ear is tested individually to establish a suitable frequency for each ear. The reference frequency may be the same for both ears, but may be different if necessary.

[0030] The reference frequency should ideally be located as close as possible to the center of the frequency range being tested. The closer the frequencies, the easier it is to compare the perceived loudness of two sounds. Age-related hearing loss typically occurs at higher frequencies, often leaving hearing ability intact in lower frequency ranges. Similarly, hearing loss caused by exposure to high volumes typically occurs at higher frequencies. Therefore, it is convenient to begin the selection of a reference frequency test with a vibrato at 800 Hz or 1200 Hz. While 1200 Hz is closer to the center of the test range than 800 Hz, hearing loss is often present at 1200 Hz, and 800 Hz may be more effective in some cases. Suitable reference frequencies are 400 Hz, 800 Hz, 1200 Hz, 1600 Hz, 2400 Hz, 3200 Hz, and 4800 Hz. In some cases, 200 Hz and 6900 Hz may be used, but accuracy is expected to be lower due to the increased distance between the reference frequency and some test frequencies.

[0031] Higher frequencies within the eligible range are typically useful for individuals suffering from a condition commonly known as cookie-bite hearing loss, where mid-range frequencies are lost while hearing at the lowest and highest frequencies is preserved. Lower frequencies can generally be used when there is significant hearing loss at 1200 Hz and are typically applied to individuals suffering from severe hearing loss resulting from aging or exposure to high volumes.

[0032] The following steps are performed individually for each ear, and vibrato is used throughout the examination. The reference frequency test begins at 60 phons and 1200 Hz. Figure 3 shows the GUI of the test panel, and the subject presses the large button on the GUI panel when they hear a sound. The order of the test frequencies is 1200 Hz, 800 Hz, 1600 Hz, 2400 Hz, 400 Hz, and 3200 Hz.

[0033] If a series of frequencies are not heard at 60 Phon, test 200 Hz, and finally 6800 Hz at 60 Phon. If no sound is heard, increase the level to 70 Phon and restart the test in the same order as above. If no sound is heard, the hearing loss is severe, and stop the test.

[0034] If a frequency is heard at a given frequency, the test is stopped at that frequency and resumed at 20 Phon, which is below the current level. If a sound is heard, the test is resumed at another level below 20 Phon. The test continues until the sound is heard at 10 Phon, and the test is stopped at that frequency, which is then designated as the reference frequency. For subjects with normal hearing at the test frequencies, the levels are in the order of 60 Phon, 40 Phon, 20 Phon, and 10 Phon, and the test is stopped when the test sound is heard at all levels. If the hearing threshold is between 20 Phon and 30 Phon, the levels are in the order of 60 Phon, 40 Phon, 20 Phon, and 30 Phon, and then the test is stopped. 30 Phon is the minimum level at which the test frequency can be heard.

[0035] If the minimum audible level for the test frequency is greater than 10 phons, the next frequency in the frequency range is tested starting from the minimum level relative to the current frequency.

[0036] If one of the subsequent frequencies in a set of frequencies is heard, that frequency is tested at a level below the minimum level of 10 Phon. The test continues until a frequency is heard at 10 Phon, and the test is stopped at that frequency and used as the reference frequency, or the test is stopped at a frequency earlier than the one heard at the minimum level and used as the reference frequency. If multiple frequencies are heard at a particular level, the first audible frequency in the set of frequencies is used as the reference frequency.

[0037] The selected reference frequencies are used to adjust the perceived loudness of other test frequencies at all volume levels; that is, there is one frequency for each ear at all volume levels, and this frequency may be the same or different for both ears.

[0038] - Step 2. Adjusting the reference frequency levels of both ears One reference frequency established for each ear in step 1 is used as a reference for adjusting all other frequencies. The perceived loudness of the other test frequencies must match the loudness of the reference frequency at all volumes.

[0039] It is essential that the perceived loudness of the reference frequency is the same in both ears at all test levels. Otherwise, the entire frequency range in which the loudness is adjusted relative to the reference will be distorted toward one ear. Furthermore, if the balance of loudness has different distortions at each level, the spatial acoustic field will shift back and forth depending on the level.

[0040] Figure 4 shows the test GUI. The left-right balance is controlled by sliders and buttons on the panel. When the reference frequency is the same in both ears, the interauricular balance is tested at each volume level used in the test. The perceived loudness levels of the left and right sounds are adjusted using gain only, without attenuation. The loudest frequency is the frequency at which hearing ability is best, and this is then used to adjust the loudness, i.e., to apply gain to sounds that sound quieter.

[0041] If the reference frequencies differ for each ear, the interauricular balance must be tested at both frequencies at each volume used in the test. For example, suppose the reference frequency for the left ear is 1200 Hz and the reference frequency for the right ear is 800 Hz. First, adjust the volume of the left ear at 800 Hz relative to the left ear's reference at 1200 Hz so that the perceived loudness of the sound in the left ear is the same at both frequencies. Similarly, adjust the volume of the right ear at 1200 Hz relative to the right ear's reference at 800 Hz. Next, adjust the 1200 Hz and 800 Hz volumes of the right ear relative to the left ear's reference at 1200 Hz, and similarly adjust the left ear relative to the right ear's reference at 800 Hz. Apply gain to all frequencies in both ears so that the quietest frequencies are adjusted relative to the loudest frequencies; that is, regardless of which ear is which, no gain is applied to the loudest frequencies, while gain is applied to all other frequencies.

[0042] - Process 3. Equal loudness testing During equal-loudness testing, the perceived loudness of each reference frequency is compared to the loudness of the test frequency at different volume levels. The loudness of the test frequency is adjusted to be the same as the test frequency.

[0043] Figure 5 shows the GUI of the test panel. The test frequency level increases when the plus bolant button is pressed and decreases when the minus button is pressed, i.e., a correction gain or loss is applied. The test frequency and / or level can be moved to the next or previous position using the next or previous button.

[0044] The appropriate level range for equal loudness testing is determined based on the measured reference frequency auditory threshold data obtained in step 1. For individuals with normal hearing or some degree of hearing loss, discomfort begins to increase at 70 phons, and 60 phons is still high but tolerable. Therefore, testing at levels above 60 phons is generally not recommended. Also, 60 phons is a very safe level that does not damage hearing. For subjects with an established reference frequency auditory threshold of less than 40 phons, 60 phons is used as the starting and maximum level. For subjects with an auditory threshold of 40 phons or 50 phons at the reference frequency, 70 phons is used as the starting and maximum level. For subjects with an auditory threshold of 60 phons or higher at the reference frequency, 80 phons is used as the starting and maximum level.

[0045] Preferably, the test frequencies are 100Hz, 200Hz, 400Hz, 800Hz, 1200Hz, 1600Hz, 2400Hz, 3200Hz, 4800Hz, 6800Hz, 9600Hz, and 12800Hz. Frequencies showing excessive hearing loss may be discarded. It is not uncommon for the higher frequency test points of 9600Hz and 12800Hz to be discarded. The criterion for discarding a test frequency is when the correction gain + level exceeds the safe sound pressure limit or the system's dynamic limit.

[0046] Equal loudness testing typically begins at the maximum volume, which is usually 60 Phon. A jump of 10 Phon levels is used as the smallest unit. If the 60 Phon test level shows a flat response, i.e., no compensation gain is needed at any frequency, 50 Phon is discarded and the next test level is set to 40 Phon. All frequencies are tested again at lower levels. If no compensation gain is still needed at any frequency, 30 Phon is skipped and the next test level is 20 Phon. If the 20 Phon test level shows a flat response, 10 Phon is tested last. If 20 Phon is not flat, 30 Phon is also tested again before going to 10 Phon last.

[0047] Generally, if no compensation gain is needed at the current level, the next level should be 10 Phon lower than the current level. A 20 Phon step is used only if no compensation gain is applied to all test frequencies. If the response is not flat at the new lowered level after the level has been lowered by 20 Phon, the next test is performed with the level raised by 10 Phon, and then the level is lowered by 10 Phon from the current level.

[0048] In the first test series, which covers all test frequencies and volumes, the correction gain is initially set to 0 dB at each frequency and level, and then adjusted up or down at each frequency, starting from 0 dB. After the first test series and testing the selected test frequencies at appropriate volumes, correction gain data can be acquired to perform a second test of the selected test frequencies and volumes. The newly acquired correction gain data is used as the starting point in the second test series. In each subsequent test series, the latest correction gain value is used as the starting point at each frequency, and this process is repeated for all subsequent test series.

[0049] After testing the first and second series, the variability between the first and second results can be evaluated. If the correction gain differs significantly from the initial 0 dB, the acquired correction gain value is likely to be slightly incorrect, and therefore it is important to repeat the test point to confirm the variability. If high variability is found, the variability of the set is calculated using priority, with the point closest to the target value as the starting point. A decision to stop is made after a suitable maximum number of tests, i.e., 5 to 10 tests, or when the variability no longer improves in the tests. Next, the average correction gain is calculated from the points with the least variability in the test series. A suitable variability threshold is 3 dB.

[0050] Only the first and second test series include all selected test frequencies at all applicable volume levels. Only the third test series includes test frequencies at volume levels where the variability exceeds the 3dB threshold. Similarly, any subsequent test series exclude test frequencies at volume levels where the variability is less than the 3dB threshold, until finally, an overall variability of less than 3dB is achieved at all test points, or the maximum limit of the number of tests is reached. The average correction gain among the test series with the least variability is calculated and used as the final gain correction.

[0051] - Step 4. Conversion from Phone to dBSPL Sound pressure levels measured in dBSPL are logarithmic units that use a sound pressure of 20 micropascals as the reference level; that is, 0 dBSPL is equal to a physical sound pressure of 20 micropascals. The ISO 226 standard provides perceived equal loudness information at different volume and frequency levels. Perceived levels are measured in phons, a unit based on the loudness of sounds that humans perceive as close. For example, two timbres at frequencies of 1 kHz and 100 Hz, both at 30 phons, are perceived as having the same loudness by a person with normal hearing ability. However, the 100 Hz timbre requires a higher physical sound pressure level to be perceived as having the same loudness as the 1 kHz timbre.

[0052] The volume levels for equal loudness testing are measured in Phon, according to the ISO 226 standard. Since Phon is a perceptual level, each frequency test point and volume level must be converted from Phon to physical sound pressure measured in dBSPL, which can be technically derived.

[0053] The ISO 226 standard includes only a finite number of numerical test points divided into one-third octave bands from 20 Hz to 12,500 Hz. The number of numerical level test points is similarly limited.

[0054] The selected frequency points for equal loudness testing do not perfectly match the frequencies in the ISO 226 standard, and the level data available in the ISO 226 standard is too coarse to be suitable for direct use. Therefore, the necessary data between the numerical test points available in the standard requires interpolation. Numerous mathematical interpolation methods are available, and in this case, the interpolated values ​​are determined by cubic spline interpolation. Both level data and frequency plane data need to be interpolated to obtain the required granularity.

[0055] During equal-loudness testing, at each test frequency and volume, the sound reproduction level is first converted from phons to physical sound pressure. Next, the sound pressure level is varied from the initial sound pressure level until it is ticked as a sound of the same loudness as the reference frequency. The compensation gain or loss added to the initial level is measured in logarithmic dB units, not in perceived phons.

[0056] As an example, we test equal loudness at 100 Hz and 20 Phon. First, we must calculate the sound pressure at the corresponding 100 Hz. Using interpolated ISO226 data, we derive the physical sound pressure level. In this case, 20 Phon at 100 Hz corresponds to approximately 48.4 dBSPL. Here, we assume that the level is too low for the subject to approach, and a positive correction gain of 10 dB is needed to achieve equal loudness. To achieve equal loudness, the test volume must be 48.4 dBSPL + 10 dB = 58.4 dBSPL. However, a level of 58.4 dBSPL is not equal to 30 Phon, and the equation 20 Phon + 10 dB = 30 Phon is incorrect. 20 phons at 100 Hz equals 48.4 dBSPL, and 30 phons at 100 Hz equals 56.8 dBSPL. The difference of 10 phons at this particular frequency and level is only 8.4 dBSPL, not 10 dB. The test volume is in phons, while the correction gain is in fundamental logarithmic units, not perceptual logarithmic units.

[0057] The conversion from Phon to dBSPL as a reference for auditory correction gain is performed in this fourth step. The measured correction gain data is interpolated from the Phon reference to the dBSPL reference. Numerous mathematical interpolation methods are available, and in this case, the interpolated dBSPL reference value is determined by cubic spline interpolation.

[0058] Any real-time processing employed to compensate for hearing loss receives physical sound pressure information from a microphone and generates physical sound pressure using a transducer. Of course, real-time applications may perform bidirectional conversion between physical sound pressure and perceptual units. However, this requires unnecessary additional computational steps, which consume both processing bandwidth and energy. Energy is typically very limited in wearable earphones, hearing aids, or similar products. Therefore, it is preferable for calculations in real-time signal processing to be based on physical sound pressure, which constructs an energy-efficient approach to achieving the desired hearing compensation. Thus, compensation gain data from equal-loudness testing based on dBSPL levels is available from the real-time signal processing unit.

[0059] Step 5. Post-processing of measured auditory data The data post-processing step aims to extend the measured auditory correction data to cover a dynamic range of at least 0 dBSPL to 110 dBSPL. Below 0 dBSPL, the same correction gain as for 0 dBSPL is used, and the level is below the auditory threshold at almost all frequencies. Similarly, levels above 110 dBSPL have the same correction gain as for 110 dBSPL, and such high levels are not amplified in any case but are attenuated by the maximum level limiter.

[0060] Equal loudness hearing tests typically provide data in increments of 10 phons, ranging from 10 phons to 60 phons at each frequency. After conversion from phons to dBSPL, data based on physical dBSPL units is available. In some cases, the available level range is smaller, but at least two, and usually four to six, measurement points are available.

[0061] First, we increase the level granularity. Interpolation can be achieved using a number of mathematical methods, in which case we perform cubic spline interpolation between raw data points, and then apply multi-order linear phase-averaged filtering to the interpolated dataset. Filtering is performed to smooth out local variations in the measured data and improve accuracy. Interpolation makes it possible to generate the desired number of data points across the obtainable dynamic range, in which case the interval between data points is 0.5 dB.

[0062] Secondly, the dynamic range is extended from below the minimum measured level to 0 dBSPL. For this purpose, the minimum level of the interpolated and filtered data is used. A linear linear polynomial derivative fitting is performed on the derivative of the lower level portion of the interpolated and filtered data. Next, the linear polynomial is used to extend the data to below the minimum measured level. Similarly, using the interpolated and filtered data at the maximum measured level, a linear linear polynomial fitting is performed on the derivative of the maximum level portion of the interpolated and filtered data. Then, the resulting linear polynomial may be used to extend the data points beyond the maximum measured level. Finally, multi-order linear phase-averaged filtering is performed on the entire extended dataset. Filtering smooths the transition between the measured and extended data, improving accuracy. Figure 6 shows 3D diagrams of the measured and extended corrected gain data, called gain sails, at seven different frequencies.

[0063] The next step involves selecting candidate correction frequencies to which correction gain can be applied. Frequencies that the equal-loudness test indicates do not require correction gain are not used and are discarded. Frequencies that the equal-loudness test indicates are also discarded because the hearing loss is severe and correction cannot be applied. Indicators include the need for excessive correction gain to restore hearing, or simply because the volume is too loud and the test is stopped. It is not uncommon for the highest frequencies of 9600Hz and 12800Hz to be problematic for people with severe age-related hearing loss, as they cannot hear these frequencies. The remaining candidate frequencies are then further analyzed to find appropriate correction frequencies.

[0064] First, the correction peaks are located within the candidate correction frequencies. In this case, a peak represents a frequency where the required correction gain is higher than that of adjacent frequencies. Once peaks are identified, a "weight" is calculated for each peak. The peak weight is the product of the distance to the next peak and the correction gain of that peak. Next, the primary peak with the largest weight is added to the selected frequency. This procedure is repeated to find further peaks among the identified peaks, and if the frequency difference between a new peak and an existing peak in the selected frequency is greater than a threshold, the new peak is sequentially added to the selected frequency. A suitable threshold is twice the distance from an existing frequency. Once all identified frequencies have been examined, the procedure stops, and the correction gain frequency is selected.

[0065] Once the appropriate frequency is selected, two optional steps are performed. If the target auditory compensation device is known, it is possible to limit the compensation gain in advance during this step, thereby eliminating or at least minimizing problems caused by acoustic feedback due to over-amplification. It is beneficial to apply the maximum compensation gain specific to each frequency that correlates with the acoustic feedback identification of the auditory compensation device to the compensation gain data at this stage.

[0066] The purpose of the final limiting step is to reduce amplification at the maximum sound pressure level. Amplification is not applied to high sound pressure levels because auditory correction is usually not necessary. As an example, Figure 6 shows the correction gain required to compensate for a subject's hearing loss. The correction does not involve amplification above approximately 80 dBSPL at all frequencies, and at 20 dBSPL, some frequencies require over 40 dB. The required correction gains shown in Figure 6 are very common, and usually, even individuals with significant hearing loss do not require gain at high sound pressure levels. In rare cases, when correction gain is present at high sound pressure levels, it is desirable to reduce the gain to avoid further hearing damage. Therefore, in this step, at levels above 90 dBSPL, the correction gain is gradually reduced to 0, and the sum of the input level and gain does not produce an output exceeding 100 dBSPL. The two threshold levels of 90 dBSPL and 100 dBSPL are selected to mitigate the risk of further hearing damage and may be adjusted to any other desired level.

[0067] - Process 6. Gain Sail Optimization Figure 6 shows a three-dimensional diagram of gainsail data for a person with cookie-bite hearing loss. The gainsail indicates the required correction gain (amplification) at frequencies 1-7 in sound pressure levels from 0 dBSPL to 120 dBSPL. In this case, the frequencies are 400 Hz, 800 Hz, 1200 Hz, 1600 Hz, 2400 Hz, 3200 Hz, and 12800 Hz. The gainsail reveals that the measured correction gain required to fully restore the subject's hearing is a collection of correction gain table data for each frequency. Looking at the gainsail, it is clear that no correction gain is needed at higher sound pressure levels, while significant amplification is needed at lower levels. Above approximately 80 dBSPL, no amplification is needed at any frequency, and at 20 dBSPL, amplification of 35 dB to 40 dB or more is needed to compensate for the hearing loss measured at intermediate frequencies.

[0068] To restore hearing loss, a real-time signal processing unit must dynamically change the filter gain at seven frequencies according to the input sound pressure level at each frequency. In this example, the real-time signal processing unit has seven bandpass filters that primarily pass each frequency to seven sound pressure level detectors, one at a time. The level detectors are then used to calculate the instantaneous required gain for the seven peak filters, again, one filter per frequency. The number of required frequencies varies depending on the case, and seven in this case is merely illustrative. In some cases, only one frequency is required, but in most cases, two to five frequencies are sufficient. Theoretically, all 12 normally obtainable frequencies could be required, but this is not typical.

[0069] Figure 8 shows the frequency response of eight corrective gain peak filters with suitable bandwidths. Gain overlap can be observed between the filters in this figure. Particularly noticeable is the overlap at relatively narrow intermediate frequencies. This narrow spacing is necessary to obtain a correction suitable for all measured hearing loss. A relatively wide bandwidth is also required; filters with wide bandwidths exhibit excellent time-domain behavior, while filters with narrow bandwidths have poor time-domain response.

[0070] Excellent time-domain behavior is required, and human hearing is very sensitive to the time-domain behavior of sound. When sounds are played on wooden or metal drums, violins, or trumpets, human hearing interprets them based on the differences in the time-domain characteristics of the sounds. It is clear that in an auditory recovery system, irregular time domains that degrade sound quality and make it difficult to hear and interpret sounds should not occur. Therefore, the corrective gain peak filter must have a wide bandwidth to maintain sound quality, and thus the response is always overlapping, as illustrated in Figure 8.

[0071] To achieve excellent tracking between dynamically applied correction gain and instantaneous sound pressure, the bandwidth of the correction peak filter and the bandpass filter of the level detector must have equivalent frequency responses. Figure 9 shows Trace 1, the frequency response of the correction peak filter at 1200 Hz, and Trace 2, the bandpass filter of the preferred detector at 1200 Hz, which overlaps with it. The bandpass filter should not be too narrow, as this will result in a poor time-domain response and therefore poor tracking of instantaneous sound pressure. Poor time-domain tracking leads to inaccurate measured instantaneous sound pressure, and therefore inaccurate applied correction gain obtained from that measured sound pressure. The bandwidth of the bandpass filter should not be too wide than that of the correction peak filter, as this will result in inaccurate sound pressure level measurements. If the bandpass filter is too wide, the measurement weight of sounds at frequencies far from the center frequency will be excessive, and therefore the gain at the center frequency will decrease. This is clearly wrong, as sound does not exist at the center frequency, and the gain applied at the center frequency becomes too small. The best results in terms of sound quality are obtained when these two filters have similar frequency responses, as illustrated in Figure 9.

[0072] If a filter with the bandwidth shown in Figure 8 is used to correct the measured hearing loss shown by the gain sail in Figure 6 without considering the gain overlap between filters, the amplification becomes excessive. The gain sail in Figure 7 shows all amplification when gain overlap is not considered. It is clear that the amplification of the combined corrected peak filter without considering gain overlap generates an excessive gain of over +90 dB at the center frequency, where it should ideally be approximately +35 dB.

[0073] In real-time signal processing type auditory correction devices, excessive gain can be corrected by feedback correction, but the feedback network always generates time-domain anomalies, which negatively affect sound quality. For example, considering that the aggregate amplification exceeds +90dB at a low input volume of 20dBBSPL, the feedback problem is clearly significant. Next, looking at Figure 6 as an example, when someone starts talking and the input volume rises significantly from silence, the gain must instantaneously drop to, for example, very low, to about 25dB. Subsequently, the gain must change to 65dB in less than 10ms, which results in significant distortion of the sound. However, not only is significant distortion produced, but the biggest problem is the gain tracking error. The gain becomes too high in the initial stages, and relatively low-volume sounds are unintentionally reproduced at very high volumes, causing a serious initial transient overshoot. Weak sounds become like gunshots in the initial stages before the gain is reduced, which is clearly a problem. Gain tracking errors always occur in such large gain adjustments over short periods of time. Furthermore, feedback requires additional real-time processing steps that inevitably consume bandwidth and energy for processing.

[0074] However, there is a better alternative to managing excess gain called gain sail optimization. By matching the bandwidth of the detector's bandpass filter with the bandwidth of the dynamic filter, it is possible to preprocess the gain table data and optimize the gain sail, thereby avoiding large distortions, gain tracking errors that cause gunshot problems, and unnecessary real-time calculations. To enable this, it is necessary to match the bandwidth of the dynamic filter and the bandwidth of the detector's bandpass filter, so that the detector can sense a volume equivalent to the gain applied by the dynamic filter. If the filter bandwidths are different, the preprocessing output will not be accurate.

[0075] Gain sail optimization is a mathematical problem that can be solved using either an analytical equation or a numerical iterative method. While an analytical solution is theoretically possible, its complexity and the enormous number of variables and equations required make this method completely impractical. Therefore, numerical methods are strongly recommended. ELHT utilizes numerical iterative methods to optimize the correction gain data.

[0076] The table data from equal-loudness hearing tests, illustrated in Figure 6, are used as targets for collective compensation gains at each frequency and associated sound pressure level. The objective of gain sail optimization is to remove the excessive compensation gain accumulated by boosting the dynamic filter at adjacent frequencies. Optimization is performed through numerical optimization of the filter gain using the least squares method.

[0077] If the best correction gain at all levels and frequencies is found through least-squares optimization, the adjusted values ​​are stored for use in the real-time signal processing unit.

[0078] Software functions, digital signal processing, and algorithms can be implemented in a variety of ways, from pure hardware implementations to pure software / firmware, or a combination of both. The DSP functions of the invention described herein use code written for a digital signal processor. The algorithms that generate the data to be input to the described ELHT and hearing recovery devices are implemented as software that runs on a personal computer. This software can, of course, be implemented to run on any computing system, such as a telephone, tablet, or other device. It can also be implemented on a dedicated target system similar to a hearing meter for a new ELHT system, or on cloud computing resources.

[0079] Specific Embodiments of the Disclosure Specific embodiments of this disclosure are disclosed below. According to this disclosure, a software unit intended for auditory evaluation and / or correction is arranged to perform a method comprising: performing a reference frequency audiometry test to establish a reference frequency for each ear of a subject; and performing an equal-loudness audiometry test for each ear, wherein the step of performing a reference frequency audiometry test includes: starting with a first frequency in a given frequency range; providing the subject with a sound at a set volume which is a minimum test volume selected from the range of -10 to 50 Phon to check for the presence or absence of an auditory response; if the subject shows an auditory response at a minimum test volume selected from the range of -10 to 50 Phon, changing the volume by the first frequency and then selecting the first frequency as the reference frequency; otherwise, selecting another frequency in a given frequency range and repeating the iterative procedure with new frequencies in a given frequency range until the software unit detects a frequency X which is the frequency at which the subject showed an auditory response at a minimum test volume selected from the range of -10 to 50 Phon or at which the subject showed a minimum volume auditory response to the set frequency being tested, and then selecting frequency X as the reference frequency.

[0080] According to one particular embodiment, the selected minimum test volume is chosen from the range of 0 to 20 phons, preferably from the range of 5 to 15 phons. As a preferred example, the minimum test volume is about 10 phons.

[0081] According to one particular embodiment, the step of changing the volume is performed by reducing the volume by a first frequency until no auditory response can be detected from the subject.

[0082] In yet another embodiment, the software unit is programmed to select frequencies closer to the center of a given frequency range than more peripheral frequencies, and preferably, if both a given high frequency and a given low frequency are equidistant from the center of the frequency range, it is programmed to select a lower frequency than a higher frequency.

[0083] Furthermore, according to another embodiment, the software unit is programmed to select new frequencies in order of priority through iterative processing. This may be done by a configured and implemented priority list.

[0084] Furthermore, according to yet another embodiment, the procedure for changing the volume is performed by at least 1 Phon in each step, preferably at least 5 Phon in each step, and more preferably at least 10 Phon in each step, until an absolute value of 10 Phon or a level at which the subject no longer shows an auditory response is reached. According to one embodiment, the procedure for changing the volume is performed by lowering the volume by at least 1 Phon in each step, preferably at least 5 Phon in each step, and more preferably at least 10 Phon in each step, until an absolute value of 10 Phon or a level at which the subject no longer shows an auditory response is reached.

[0085] Furthermore, according to one embodiment, a software unit is positioned to perform a left-right channel level adjustment test, which includes adjusting the perceived loudness of a reference frequency between the subject's two ears.

[0086] According to one embodiment, the step of adjusting the perceived loudness of a reference frequency between the subject's ears is performed before the equal-loudness hearing test for each ear.

[0087] In yet another embodiment, a software unit is positioned to perform a left / right channel level adjustment test between the step of performing a reference frequency hearing test and the step of performing an equal-loudness hearing test. Furthermore, in one embodiment, the software unit is positioned to perform a left / right channel level adjustment test at all set volume levels. In yet another embodiment, the step of performing an equal-loudness hearing test for each ear is performed by comparing the perceived loudness of a reference frequency, which is one for each ear, with a plurality of test frequencies.

[0088] Furthermore, the software unit may be configured to perform an equal-loudness hearing test by providing a frequency different from a set reference frequency and comparing the response from the subject at several appropriate sound pressure levels that exceed the subject's auditory threshold. According to yet another embodiment, the software unit is configured to perform a first series of tests for equal-loudness hearing by providing the subject with a sequence of different volume levels and checking whether a compensation gain is needed at each volume level. Furthermore, according to yet another embodiment, the sequence of different volume levels in the first test series is provided by decreasing the volume until a volume requiring a compensation gain is found, preferably the volume subsequently tested is higher than the volume established as requiring a compensation gain.

[0089] According to one embodiment, the software unit is configured to run a second test series to obtain a new corrected gain value, using the gain correction value established from a first test series for different volume levels provided to the subject, preferably the established corrected gain value being used iteratively in subsequent test series. Furthermore, the software unit may be configured to evaluate the variability of the corrected gain value between different test series. A maximum threshold for variability may also be used as an input for deciding to stop the test for a particular volume level, preferably the maximum threshold being 3 dB. According to yet another embodiment, the software unit is configured to calculate the average corrected gain for the minimum variability portion in the performed test series and set it as the final gain correction. In addition, the software unit is preferably configured to run equal-loudness auditory tests using vibrato, preferably with vibrato modulation in the range of 1 to 10%, and preferably with modulation frequencies in the range of 5 Hz to 30 Hz.

[0090] As can be seen from the above, according to this disclosure, the software unit may preferably be configured to collect data obtained from the subject. Furthermore, according to yet another embodiment, the software unit may be configured to discard one or more tested frequencies, preferably one or more tested frequencies based on the subject's hearing loss to the one or more tested frequencies.

[0091] Furthermore, according to one embodiment of the present disclosure, a software unit is arranged to set an SPL range for equal-loudness audiometry, preferably in the range of 10 to 70 phons, and equal-loudness audiometry is performed by changing the volume, preferably by at least 1 phon in each step, more preferably by at least 5 phons in each step, and even more preferably by at least 10 phons in each step. Furthermore, according to one embodiment, the step of setting an SPL range for equal-loudness audiometry is performed by reducing high values ​​within the set range.

[0092] According to one embodiment, the software unit may be located in a computer unit or mobile terminal such as a telephone or tablet, or as part of an embedded system such as a dedicated product.

[0093] Furthermore, according to yet another embodiment, a software unit or connected data computing software is arranged to convert acquired data from Phon to absolute dBSPL values ​​by interpolating measured corrected gain data from a Phon reference to a dBSPL reference. In this regard, any arbitrary setup with different connected software is possible according to the Disclosure, as described below. According to one embodiment of the Disclosure, a software unit or connected data computing software is arranged to extend the dynamic range to below the maximum measured level and / or above the minimum measured level, preferably using a linear polynomial, more preferably by multi-order linear phase FIR average filtering. In this regard, the application may be provided by a telephone or computer unit and may further be connected to another program where additional calculations are performed, such as a cloud.

[0094] Furthermore, according to one embodiment, a software unit or connected data computing software identifies and selects candidate correction frequencies to which a correction gain can be applied by searching for peak frequencies where the required correction gain is higher than adjacent frequencies, then preferably calculates the weights of all identified peaks based on the distance to adjacent peaks and the correction gain of a particular peak, then more preferably identifies the principal peak having the largest weight and places it to be added to the selected frequency.

[0095] Furthermore, according to yet another embodiment, a software unit or connected data computing software is arranged to perform gain sail optimization, which includes compensation for adjacent filter gain contributions where a collection of amplifications from adjacent filters exists. Gain sail optimization may be performed in a different location according to this disclosure, such as an additional software unit or main software. Furthermore, according to one embodiment, gain sail optimization includes removing excess correction gain accumulated by adjacent frequency-corrected peak filter boost, preferably by performing least-squares numerical optimization of the filter gain.

[0096] As described above, the software unit relating to this disclosure may be part of a system. In this regard, according to one embodiment, a system comprising the software unit relating to this disclosure is provided, wherein the software unit is connected to a digital signal processing (DSP) unit, or to another software unit or device comprising a digital signal processing (DSP) unit, and the digital signal processing (DSP) unit is arranged to process and compensate for data acquired from collected subjects.

[0097] According to one embodiment, the digital signal processing (DSP) unit comprises one or more dynamic filters for compensating data, preferably a plurality of dynamic filters for compensating data, preferably 2 to 20 dynamic filters for compensating data, and preferably each dynamic filter operates with a gain that changes dynamically depending on the dynamically changing input signal level at the filter frequency. According to yet another embodiment, the digital signal processing (DSP) unit is arranged for digital signal processing of amplified data and filter center frequencies, and preferably comprises one or more filter blocks. According to another embodiment, the digital signal processing (DSP) unit comprises one or more dynamic filters for compensating data, and each dynamic filter has a center frequency in the range of 100 Hz to 12.8 kHz.

[0098] In yet another embodiment, the one or more dynamic filters operate with a gain that changes dynamically depending on a dynamically changing input signal level at a particular filter frequency. Preferably, the one or more filter blocks included in a digital signal processing (DSP) unit include a bandpass filter, a sound pressure detector, and a dynamic filter. In yet another embodiment, the bandpass filter is configured to remove frequencies, and the detector is configured to measure the signal level at the dynamic filter frequency and suppress sound signals present at other frequencies.

[0099] Furthermore, according to one embodiment, the bandpass filter has a low order, preferably second-order, and more preferably a Q value of less than 1. Furthermore, according to yet another embodiment, the bandpass filter and the dynamic filter have the same bandwidth.

[0100] Furthermore, according to one embodiment, the one or more filter blocks also include a gain table that uses table data from an algorithm or the like to convert the current input level into a gain setting for a dynamic filter.

[0101] In yet another embodiment, table amplification data and filter center frequencies are imported from a software unit algorithm to a digital signal processing (DSP) unit, which comprises one or more filter blocks that provide dynamically different amplification, each handling a unique frequency range.

[0102] In yet another embodiment, multiple filter blocks are used in a digital signal processing (DSP) unit.

Claims

1. A software unit intended for auditory evaluation, The process involves performing a reference frequency hearing test to establish a reference frequency for each ear of the subject, The process involves performing equal-loudness hearing tests on each ear, It is arranged to perform a method that includes, The step of performing the aforementioned reference frequency hearing test is: Starting from a first frequency within a given frequency range, In order to confirm the presence or absence of an auditory response, a sound is provided to the subject at a set volume, which is the minimum test volume selected from the range of -10 to 50 Phon. If the subject shows the auditory response at the minimum test volume selected from the range of 10 to 50 Phon, the volume is changed according to the first frequency, and then the first frequency is selected as the reference frequency. Otherwise, the process involves selecting another frequency within the given frequency range and repeating the iterative procedure with new frequencies within the given frequency range until the software unit detects a frequency X which is the frequency at which the subject showed the auditory response at the minimum test volume selected from the range of -10 to 50 Phon, or the frequency at which the subject showed the minimum volume auditory response to the set frequency being tested, and then selecting frequency X as the reference frequency. including, Software unit.

2. The software unit of claim 1, wherein the selected minimum test volume is selected from the range of 0 to 20 Phon, preferably from the range of 5 to 15 Phon.

3. The software unit according to claim 1 or 2, wherein the step of changing the volume is performed by lowering the volume by the first frequency until the auditory response can no longer be confirmed from the subject.

4. The aforementioned software unit is It is programmed to select a frequency that is closer to the center of the given frequency range than to the surrounding frequencies. Preferably, if both a high frequency and a low frequency are equidistant from the center of the frequency range, the system is programmed to select the lower frequency over the higher frequency. A software unit according to any one of claims 1 to 3.

5. A software unit according to any one of claims 1 to 5, which is programmed to select a new frequency in order of priority through iterative processing.

6. The software unit of any one of claims 1 to 5, wherein the procedure for changing the volume is performed by at least 1 Phon in each step, preferably at least 5 Phon in each step, and more preferably at least 10 Phon in each step, until an absolute value of 10 Phon or a level at which the subject no longer exhibits the auditory response is reached.

7. The software unit of claim 6, wherein the procedure for changing the volume is performed by reducing the volume by at least 1 Phon in each step, preferably at least 5 Phon in each step, and more preferably at least 10 Phon in each step, until the volume reaches an absolute value of 10 Phon or a level at which the subject no longer exhibits the auditory response.

8. The software unit according to any one of claims 1 to 7, wherein the software unit is arranged to perform a left-right channel level adjustment test, which includes adjusting the perceived loudness of the reference frequency between the two ears of the subject.

9. The software unit of claim 8, wherein the step of adjusting the perceived loudness of the reference frequency between the ears of the subject is performed before the equal loudness hearing test for each ear.

10. The software unit according to claim 8 or 9, wherein the software unit is positioned between the step of performing the reference frequency hearing test and the step of performing the equal loudness hearing test to perform the left and right channel level adjustment test.

11. The software unit according to any one of claims 8 to 10, wherein the software unit is configured to perform the left and right channel level adjustment check at all set volume levels.

12. The software unit according to any one of claims 1 to 11, wherein the step of performing an equal-loudness hearing test for each of the ears is performed by comparing the perceived loudness of one reference frequency for each of the ears with a plurality of test frequencies.

13. The software unit according to any one of claims 1 to 12, wherein the software unit is arranged to perform an equal-loudness hearing test by providing a frequency different from the set reference frequency and comparing the response from the subject at a plurality of appropriate sound pressure levels that exceed the subject's hearing threshold.

14. The software unit according to any one of claims 1 to 13, wherein the software unit is arranged to perform a first series of tests of the equal loudness hearing test by providing the subject with a sequence of different volume levels and checking whether the correction gain is necessary at each of the volume levels.

15. The sequence of different volume levels in the first test series is provided by reducing the volume until the volume level requiring the correction gain is found. Preferably, the volume subsequently tested is higher than the volume at which the correction gain was established to be necessary. The software unit according to claim 14.

16. The software unit is configured to use the gain correction values ​​established from the first test series for different volume levels provided to the subject and to perform a second test series to obtain new corrected gain values. Preferably, the established correction gain value is used iteratively in the next series of tests. The software unit according to claim 14 or 15.

17. The software unit according to claim 16, wherein the software unit is arranged to evaluate the variability of the correction gain value between different inspection series.

18. The aforementioned maximum threshold for variability is used as an input for determining when to stop the inspection for a particular volume level. Preferably, the maximum threshold is 3 dB. The software unit according to claim 17.

19. The software unit according to claim 17 or 18, wherein the software unit is configured to calculate the average correction gain of the minimum variation portion in the performed test series and set it as the final gain correction.

20. The software unit according to any one of claims 1 to 19, wherein the software unit is arranged to perform the equal loudness hearing test using vibrato, preferably using vibrato modulation in the range of 1 to 10%, and preferably using a modulation frequency in the range of 5 Hz to 30 Hz.

21. The software unit is configured to collect data obtained from the subject, according to any one of claims 1 to 20.

22. The software unit is arranged to discard one or more tested frequencies. Preferably, the one or more test frequencies are based on the subject's hearing loss to the one or more tested frequencies. A software unit according to any one of claims 1 to 21.

23. The software unit is preferably arranged to set the SPL range for the equal loudness hearing test, which is within the range of 10 to 70 phons. The aforementioned equal loudness hearing test is performed by changing the volume, preferably by at least 1 phon in each step, more preferably by at least 5 phons in each step, and even more preferably by at least 10 phons in each step. A software unit according to any one of claims 1 to 22.

24. The software unit of claim 23, wherein the step of setting the SPL range for the equal loudness hearing test is performed by reducing a high value within the set range.

25. The software unit is located in any of claims 1 to 24, wherein the software unit is placed in a computer unit or mobile terminal such as a telephone or tablet, or as part of an embedded system such as a dedicated product.

26. The software unit or connected data computing software is configured to convert acquired data from Phon to absolute dBSPL values ​​by interpolating measured correction gain data from a Phon reference to a dBSPL reference, according to any one of claims 1 to 25.

27. The software unit or the connected data computing software is arranged to extend the dynamic range to below the maximum measured level and / or above the minimum measured level, preferably using a first-order polynomial, and more preferably by multi-order linear phase FIR averaged filtering, according to claim 26.

28. The software unit or the connected data computing software identifies and selects candidate correction frequencies to which the correction gain can be applied by searching for peak frequencies where the required correction gain is higher than adjacent frequencies. Next, preferably, the weights of all identified peaks are calculated based on the distance to adjacent peaks and the correction gain of a particular peak. Next, more preferably, the principal peak having the greatest weight is identified and placed to be added to the selected frequency. A software unit according to claim 26 or 27.

29. The software unit or the connected data computing software is arranged to perform gain sail optimization, which includes compensation for adjacent filter gain contributions, where a set of amplifications from adjacent filters exists, according to any software unit of claims 26 to 28.

30. The software unit of claim 29, wherein the gain sail optimization preferably includes removing the excess correction gain accumulated by adjacent frequency corrected peak filter boost by performing least squares numerical optimization of the filter gain.

31. A software unit comprising any one of claims 1 to 30, The software unit is connected to a digital signal processing (DSP) unit, or to another software unit or device equipped with the digital signal processing (DSP) unit. The digital signal processing (DSP) unit is arranged to process and compensate for the data acquired from the subjects that have been collected. system.

32. The digital signal processing (DSP) unit includes one or more dynamic filters for data compensation. Preferably, the system includes multiple dynamic filters to compensate for the data, Preferably, the system includes 2 to 20 of the dynamic filters to compensate for the data. Preferably, each of the dynamic filters operates with a gain that changes dynamically depending on the dynamically changing input signal level at the filter frequency. The system according to claim 31.

33. The aforementioned digital signal processing (DSP) unit is: Arranged for digital signal processing of amplified data and filter center frequency, Preferably, comprising one or more filter blocks, The system according to claim 31 or 32.

34. The aforementioned digital signal processing (DSP) unit is: The system comprises one or more of the aforementioned dynamic filters for compensating for data, Each of the aforementioned dynamic filters has a center frequency in the range of 100 Hz to 12.8 kHz. The system according to claim 32 or 33.

35. The system according to claims 32 to 34, wherein one or more of the dynamic filters operate with a gain that changes dynamically depending on the dynamically changing input signal level at a specific filter frequency.

36. The system according to claims 33 to 35, wherein the one or more filter blocks comprises a bandpass filter, a sound pressure detector, and a dynamic filter.

37. The aforementioned bandpass filter is arranged to remove frequencies, The detector is positioned to measure the signal level at the dynamic filter frequency and to suppress the sound signal present at another frequency. The system according to claim 36.

38. The system of claim 36 or 37, wherein the bandpass filter preferably has a low order, with a Q value of less than 1.

39. The system according to claims 36 to 38, wherein the bandpass filter and the dynamic filter have the same bandwidth.

40. The system according to any one of claims 33 to 29, wherein one or more filter blocks also include a gain table using table data from an algorithm or the like, and converts the current input level to the gain setting of the dynamic filter.

41. The table amplification data and filter center frequency are imported from the software unit algorithm to the digital signal processing (DSP) unit. The DSP unit comprises one or more filter blocks that provide dynamically different amplification, each handling a unique frequency range. A system according to any one of claims 33 to 40.

42. The system according to claims 33 to 41, wherein a plurality of the filter blocks are used in the digital signal processing (DSP) unit.

Images