METHOD FOR ACCURATELY ESTIMATING A PURE TONE THRESHOLD USING AN UNREFERENCED AUDIO-SYSTEM

Abstract

Method for estimating a pure tone hearing threshold or a pure tone audiogram of a person with a non-calibrated audio-system, comprising the steps of performing a supra-threshold test at least over a portion of the audible frequency spectrum, wherein the audible frequency spectrum ranges particularly from 16 Hz to 20.000 Hz, wherein the supra-threshold test is performed on an unreferenced audio-system at a first relative sound level relative to a predefined output level of the unreferenced audio-system, determining the result of the supra-threshold test for at least a portion of the audible frequency spectrum, wherein the results of the supra-threshold test are provided particularly relative to the predefined output level of the unreferenced audio-system, determining from the progression of the determined result of the supra-threshold test at least one absolute pure tone threshold, wherein the at least one absolute pure tone threshold is provided in absolute physical units, particularly in decibels hearing level or decibels sound pressure level.

Claims

1. A method for estimating a pure tone hearing threshold or a pure tone audiogram of a person from results of a supra-threshold test with an unreferenced audio-system on a mobile device, the method comprising: performing, at a mobile device, a supra-threshold test at least over a portion of an audible frequency spectrum with a person, wherein the audible frequency spectrum ranges particularly from 16 Hz to 20,000 Hz, wherein the supra-threshold test is performed on an unreferenced audio-system at a first sound level relative to a predefined output level of the unreferenced audio-system, determining, at the mobile device, a result of the supra-threshold test for at least a portion of the audible frequency spectrum, wherein the result of the supra-threshold test is provided particularly relative to the predefined output level of the unreferenced audio-system, determining, at the mobile device, from the result of the supra-threshold test at least one absolute pure tone threshold, wherein the at least one absolute pure tone threshold is provided in absolute physical units, particularly in decibels hearing level or in decibels sound pressure level.

2. The method according to claim 1, wherein the result of the supra-threshold test comprises or is a set of characterizing parameters, wherein the characterizing parameter set comprises at least one parameter, wherein the parameter is configured to describe a progression, a shape feature, a gradient and/or a shape of the particularly graphical representation of the result of the supra-threshold test or wherein the characterizing parameter set is or comprises principal components of the result of the supra-threshold test, wherein the characterizing parameter set particularly comprises not more than 5 principal components of the results of the supra-threshold test.

3. The method according to claim 1, wherein the at least one absolute pure tone threshold is determined by a regression function, wherein the result of the supra-threshold test, particularly in form of a characterizing parameter set, of the performed supra-threshold test is submitted to the regression function, wherein the regression function is configured to determine from the result of the submitted supra-threshold test the at least one pure tone threshold.

4. The method according to claim 3, wherein the regression function is determined by: providing a training set comprising a plurality of results of the supra-threshold test, each particularly in form of a characterizing parameter set, wherein the results of the supra-threshold test are particularly estimated from a plurality of people, wherein to each results of the supra-threshold test of the training set at least one absolute pure tone threshold is associated, determining from the training set a regression function between the results of the supra-threshold test and the associated at least one absolute pure tone threshold by regression analysis.

5. The method according to claim 3, wherein the regression function is a multivariate linear regression function, wherein particularly the variables of the regression function are the characterizing parameters, and particularly wherein the coefficients for the regression function have been determined from a training set.

6. The method according to claim 1, wherein the at least one absolute pure tone threshold is determined by: providing database of a plurality of results of the supra-threshold test, particularly acquired from a plurality of persons, wherein at least one absolute pure tone threshold is associated to each provided results of the supra-threshold test, determining a similar results of the supra-threshold test from the database based on a predefined similarity metric between the results of the supra-threshold test, assigning the at least one absolute pure tone threshold associated to the similar results of the supra-threshold test to the results of the supra-threshold test from the performed supra-threshold test.

7. The method according to claim 1, wherein the supra-threshold test comprises or is a psychometric tuning curve test and/or comprises or is a temporal fine structure test, and/or comprises or is a temporal masking curve test.

8. The method according to claim 1, wherein a plurality of supra-threshold tests are performed on the unreferenced audio-system, wherein the supra-threshold tests are particularly performed at different sound levels and/or in different portions of the frequency spectrum.

9. The method according to claim 1, wherein the supra-threshold test is a psychometric tuning curve test, wherein the psychometric tuning curve test is particularly measured for signal tones at frequencies of 500 Hz, 1 kHz, 2 kHz and/or 4 kHz, and particularly at a sound level of 20 dB SL, 30 dB SL, and/or 40 dB SL, wherein particularly a masking signal of each the signal tone sweeps from 60% of the signal tone frequency to 140% of the signal tone frequency.

10. The method according to claim 1, wherein before the supra-threshold test is performed, the method comprising: performing a pure tone threshold test with the unreferenced audio-system, wherein the sound level for each signal tone is referenced to the predefined output level of the unreferenced audio-system, determining pure tone thresholds, particularly a pure tone audiogram from the pure tone threshold test, wherein the thresholds are provided relative to the predefined output level of the unreferenced audio-system, wherein the first sound level of the supra-threshold test is higher by a predefined value than at least one of the determined pure tone thresholds provided relative to the predefined audio level, and/or wherein after the at least one absolute pure tone threshold is determined, the pure tone thresholds estimated before the supra-threshold test are referenced against the at least one absolute pure tone threshold.

11. The method according to claim 1, wherein the predefined output level is estimated by providing information about the hardware features or components of the audio-system, particularly prior the supra-threshold test or prior the pure tone threshold test is performed.

12. The method according to claim 1, wherein a second supra-threshold test is performed with a second sound level relative to the predefined output level of the unreferenced audio-system, and wherein the at least one pure tone threshold is determined from the result of the second supra-threshold tests performed at the first and the second sound level, particularly by regression function, more particularly by a multivariate linear regression function.

13. The method according to claim 1, wherein the determined results of the supra-threshold test is a function of at least one provided frequency, wherein the progression of the results of the supra-threshold result is particularly a shape, a slope or a specific feature of the function particularly when displayed graphically, such as a width of a locally v-shaped or locally parabolic-shaped function, a local minimum, and/or the steepness of a locally v-shaped or locally parabolic-shaped function.

14. The method according to claim 1, wherein a plurality of different supra-threshold tests are performed and for each of obtained result of the supra-threshold tests or for the combination of the obtained results of the supra-threshold tests the at least one absolute pure tone threshold is determined.

15. The method according to claim 1, wherein the at least one absolute pure tone threshold is determined in combination with results that have been determined by the method according to the invention, but on a different unreferenced audio-system.

Description

[0153] Further features and advantages of the invention shall be described by means of a detailed figure description, wherein features disclosed in the figure section can also be used in combination with the claimed subject matter.

[0154] FIG. 1 shows an illustration of a PTC measurement. A signal tone 102 is masked by a masker signal 105 particularly sweeping 103 through different frequencies in the proximity of the signal tone 102. The test person is indicating at which sound level he hears the signal tone for each masker signal. The signal tone and the masker signal are well within the hearing range of the person.

[0155] The diagram shows in the x-axis the frequency and on the y-axis the audio level or intensity in arbitrary units.

[0156] While a signal tone 102 that is constant in frequency and intensity 101 is played to the person a masker signal 105 slowly sweeps 103 from a frequency lower to a frequency higher than the signal tone 102. The rate of sweeping 103 is constant or can be controlled by the test person or the operator. The goal for the test person is to hear the signal tone 102. When the test person is not hearing the signal tone 102 anymore (which is for example indicated by the user by releasing a pushbutton) the masker signal is intensity is reduced 104 to a point where test person starts hearing the signal tone 102 (which is for example indicated by the user by pressing the push button). While the masker signal tone 105 is still sweeping 103 upwards in frequency, the intensity of the masker signal 105 is increased 104 again, until the test person does not hear the signal tone 102 anymore. This way, the masker signal intensity oscillates 106 around the hearing level 107 (as indicated by the solid line) of the test person with regard to the masker signal frequency and the signal tone.

[0157] This hearing level 107 is well established and well known for people having no hearing loss. Any deviations from this curve indicate a hearing loss (see for example FIG. 2).

[0158] FIG. 2 shows the test results acquired on a calibrated setup in order to generate a training set for the method according to the invention. Therefore, the acquired PTC tests 200 can be given in absolute units such as dB HL. However, this is not crucial for the further evaluation.

[0159] In the present example, four PTC tests 200 at different signal tone frequencies 201 (500 Hz, 1 kHz, 2 kHz and 4 kHz) and at three different sound levels (40 dB HL, 30 dB HL and 20 dB HL; indicated by the thickness of the lines; the thicker the line the lower the signal tone level) for each signal tone have been performed. Therefore, at each signal tone frequency, there are three PTC curves. The PTC curves each are essentially v-shaped.

[0160] Dots below the PTC curves indicate the results from a calibratedand thus absolutepure tone threshold test performed with the same person.

[0161] On the upper panel, the PTC results and pure tone threshold test results acquired from a normal hearing person are shown, wherein on the lower panel, the same tests are shown for a hearing impaired person.

[0162] In the example shown, a training set comprising 20 persons, both normal hearing and hearing impaired persons, has been acquired.

[0163] In FIG. 3 a summary of the PTC test results of the training set are shown 300. The plots are grouped according to single tone frequency and sound level resulting in 12 panels.

[0164] In each panel the PTC results are grouped in 5 groups (indicated by different line styles), according to their associated pure tone threshold test result. In some panels pure tone thresholds were not available, so these groups could not be established.

[0165] The groups comprise the following pure tone thresholds indicated by line colour: thin dotted line: >55dB, thick dotted line: >40 dB,: dash-dot line >25 dB, dashed line: >10 dB and continuous line: >5 dB

[0166] The PTC curves have been normalized relative to signal frequency and sound level for reasons of comparison. Therefore, the x-axis is normalized with respect of the signal tone frequency. The x-axes and y-axes of all plots show the same range.

[0167] As can easily be discerned across all graphs, elevations in threshold gradually coincide with wider PTCs, i.e. hearing impaired (HI) listeners have progressively broader tuning compared to normal hearing (NH) people.

[0168] This qualitative observation can be used for quantitatively determining at least one pure tone threshold from the shape-features of the PTC.

[0169] Modelling of the data is realised using a multivariate linear regression function of individual pure tone thresholds against corresponding PTCs across users, with separate models fit for each experimental condition (i.e. for each signal tone frequency and sound level).

[0170] To capture the dominant variabilities of the PTCs across users and in turn reduce dimensionality of the predictors, i.e. to extract a characterizing parameter set PTC traces are subjected to a principle component analysis (PCA). Including more than the first five PCA components does not improve predictive power. FIGS. 4 and 5 summarise the fitted models' threshold predictions. Across all users and conditions, the standard absolute error of estimation amounted to 4.8 (1.7) dB, 89% of threshold estimates were within standard 10 dB variability. FIG. 6 plots regression weights across PTC masker frequency and indicates that mostly low-, but also high-frequency regions of a PTC trace are predictive of corresponding thresholds.

[0171] Thus, with the such generated regression function it is possible to determine an absolute pure tone threshold from an uncalibrated audio-system, as particularly the shape-feature of the PTC can be used to conclude form a PTC of unknown absolute sound level to the absolute pure tone threshold.

[0172] FIG. 4 shows the PTC-predicted 400 vs. true audiometric pure tone thresholds across all users and experimental conditions (marker size indicates the PTC signal level). Dashed (dotted) lines represent unit (double) standard error of estimate.

[0173] FIG. 5 shows a histogram of the differences between true and predicted pure tone thresholds, including an approximate normal distribution (smooth solid line).

[0174] FIG. 6 shows regression weights of linear model across PTC signal frequencies (=subplots) and signal levels (=line thickness) on a frequency normalized plot.

[0175] FIG. 7 shows a flow diagram of the method according to the invention. First, a training phase is initiated, where on a calibrated setup, PTC data are collected (step a.i).

[0176] In a.ii these data are pre-processed and then analysed for PTC features (step a.iii).

[0177] The training of the classifier (step a.v) takes the PTC features (also referred to as characterizing parameters) as well as related pure-tone thresholds (step a.iv) as input.

[0178] The actual prediction phase starts with step b.i, in which PTC data are collected on an uncalibrated setup. These data are pre-processed (step b.ii) and then analysed for PTC features (step b.iii).

[0179] The classifier (step c.i) using the setup it developed during the training phase (step a.v) predicts at least one pure-tone threshold (step c.ii) based on the PTC features of an uncalibrated setup.