TWM636702U - Self-fitting hearing compensation device with real ear measurement and hearing apparatus with the self-fitting hearing compensation device - Google Patents

Abstract

Provided is a self-fitting hearing compensation device with real ear measurement, including a first transducer, which receives a first test signal from a device and converts the first test signal into a first electrical signal; a first hearing compensation module, which is connected to the first transducer and a wireless transceiving module, and performs gain compensation on the first electrical signal; a second transducer, which is connected to the first hearing compensation module, converts the gain-compensated first electrical signal into sound, and transmits the sound into the ear canal; and a third transducer, which is connected to the wireless transceiving module and synchronously converts the sound transmitted in the ear canal into a second electrical signal, so as to transmit the second electrical signal to the device through the wireless transceiving module and a wireless transmission network. The device calculates an energy distribution of the second electrical signal in each frequency band, and compares an error between the energy distribution and a target gain and a hearing threshold by means of a second hearing compensation module. If the error does not meet an error target, the device quantizes the error to generate corrected filter parameters by a compensation gain conversion model. The corrected filter parameters are transmitted to the first hearing compensation module via the wireless transmission network to perform hearing gain compensation. In addition, the present disclosure also provides a hearing apparatus with the self-fitting hearing compensation device.

Description

Self-adjusting auditory compensation device with real ear measurement and hearing equipment with self-adjusting auditory compensation device

The invention relates to an auditory compensation technology, in particular to a self-fitting auditory compensation device with real ear measurement (REM) analysis and hearing equipment with the self-adjustable auditory compensation device.

According to statistics, there are more than 120,000 hearing-impaired persons in Taiwan with certificates of physical and mental disabilities. Hearing problems will lead to invisible or significant problems in various important aspects of life, such as language communication, occupational adaptation, social participation, school learning, and life safety. obstacles.

Furthermore, Taiwan is about to enter a super-aged society, and hearing loss ranks among the top three chronic diseases among the elderly. At present, due to the advancement of modern hearing aids and hearing aids technology on the market, the negative impact and burden of hearing impairment on individuals, families, communities, and even the entire society can be greatly improved.

However, the technology of modern hearing aids and hearing aids still requires a real ear analyzer to be adjusted. The known real ear analyzer is equipped with a probe tube, which has a first microphone and a second microphone respectively. The first microphone is responsible for collecting the sound near the opening of the ear canal , while the second microphone is responsible for collecting sounds near the eardrum. When performing real ear test analysis, first insert the probe tube into the ear canal, and the tip of the probe tube is about 5mm away from the tympanic membrane, and test the sound changes in the ear canal when not wearing a hearing aid and after wearing a hearing aid, and then get the real ear insertion response REIR. Specifically, the operating steps of the known real ear test are: 1) When no hearing aid is placed at the opening of the ear canal, that is, when the opening of the ear canal is open, the sound field emits sound (including all frequencies and the same sound pressure level), the first microphone and The sound pressure level difference curve recorded by the second microphone is called real ear unaided response REUR (also called real ear unaided gain REUG); and 2) when a hearing aid is placed at the ear canal, the sound field emits sound, the first The sound pressure level difference curve recorded by the microphone and the second microphone is called the real ear aided response REAR (also known as the real ear aided hearing gain REAG), but because the real ear test must be detected by an instrument in the hearing room and needs to be tested through hearing It is less efficient and less timely to carry out with the assistance of professionals.

In addition, the measured real ear response is often inconsistent with the expected results of the fitting software, mainly because the acoustic characteristics (such as resonance, volume, impedance, etc.) The "average ear" information of the person is different. When the real ear test is performed, the unique ear canal characteristics of the hearing-impaired person will be reflected, resulting in some errors. Furthermore, the acoustic parameters of hearing aids for hearing-impaired persons differ, for example, the size of the air hole or the depth of the eardrum. Therefore, real ear testing requires additional gain adjustments to match the specified or expected target gain.

Additionally, insertion gain measurements are a common method for verifying hearing aid performance characteristics. However, as mentioned above, the insertion gain has many limitations when debugging hearing aids, resulting in errors.

Based on the above-mentioned reasons, how to provide a real ear measurement and analysis that does not require a real ear analyzer, a probe transducer (that is, a probe microphone (probe microphone)), and does not need to be limited to a professional hearing space (such as an audiology room) , and does not require the assistance of hearing professionals (such as professional tuners), the hearing compensation device and the hearing equipment with the hearing compensation device that can effectively solve the above problems can provide accurate and real-time , Automated and customized hearing aids for users (especially hearing-impaired patients) (such as hearing aids, hearing aids or earphones with hearing aid function, hearing equipment such as glasses, ANC earphones or TWS earphones, etc.), and have become the industry's urgent issues to be resolved.

In order to solve the aforementioned known technical problems or provide related effects, this invention provides a self-adjusting auditory compensation device with real ear measurement, which includes: a first transducer for receiving a first test signal from a device, and converting the first test signal into a first electrical signal; the hearing compensation module is connected to the first transducer and the wireless transmission and receiving module, and performs gain compensation on the first electrical signal; the second conversion The transducer is connected to the hearing compensation module, converts the first electrical signal after gain compensation into sound, and transmits the sound into the ear canal; and the third transducer is connected to the wireless transmission A receiving module, the third transducer synchronously converts the sound transmitted in the ear canal into a second electrical signal, so as to transmit the second electrical signal to the wireless transmission receiving module through the wireless transmission network The device, wherein the device calculates the energy distribution of the second electrical signal in each frequency band, and compares the error of the energy distribution with the target gain and hearing threshold, and if the error does not meet the error target, the device calculates the error quantified by complementing After the modified filter parameters are generated by the compensation gain conversion model, the modified filter parameters are sent to the hearing compensation module through the wireless transmission network for hearing gain compensation.

The invention also provides a hearing device with the above-mentioned self-adjusting hearing compensation device.

In one embodiment, the hearing compensation module is set in an active noise reduction chip or a chip of a digital signal processing circuit.

In one embodiment, the modified filter parameters are gain compensation filter parameters of ANC or gain compensation parameters of a digital signal processing circuit.

In one embodiment, the filter parameter of the gain compensation of the ANC is an audio gain compensation filter unit parameter. In other words, the gain compensation unit of the ANC technology is an audio gain compensation filter unit of the ANC technology, such as an SZ or APT filter, and the filter parameters of the audio gain compensation filter unit (such as SZ or APT filter parameters).

In one embodiment, the present invention includes a storage module, wherein the device stores the corrected filter parameters to the storage module if the error meets an error target.

In one embodiment, the device stores the original filter parameters in a device with audio processing capability, wherein the device has a hearing compensation module for hearing gain compensation.

In one embodiment, the hearing compensation module in the device is based on the real-time customized audiogram or audiogram obtained by the hearing compensation device in the current real environment, and combines optimization methods and The loss function automatically searches the optimal filter parameter value generated by the complex set of parameters of the complex filter as the original filter parameter, but the present invention is not limited thereto.

In one embodiment, the wireless transmitting and receiving module of the present invention receives the second test signal from the device through the wireless transmission network to perform hearing gain compensation. In addition, the first test signal is transmitted in the air, and the second test signal is transmitted through wireless communication.

In one embodiment, if the error still fails to meet the error target, the device re-quantizes the error and transmits the quantized error and auditory dynamic range application optimization parameters to the compensation gain conversion model, so that the After the compensation gain conversion model generates another modified filter parameter, the another modified filter parameter is sent to the hearing compensation module through the wireless transmission network to perform the hearing gain compensation.

In one embodiment, the self-adjusting auditory compensation device with real ear measurement and the hearing equipment are used for audiometry in a non-audiometry room environment.

In one embodiment, the self-adjusting auditory compensation device with real ear measurement is set in a hearing aid with active noise reduction or digital signal processing circuit, and the self-adjusting auditory compensation device is not an auditory device in the hearing examination room. In-room earphones are used without the assistance of a hearing professional; in another embodiment, the hearing device is applied to a hearing aid with active noise reduction or digital signal processing circuitry.

In one embodiment, the self-adjusting auditory compensation device with real ear measurement and the hearing equipment are automatically, real-time and/or synchronized through the application of the device in combination with the compensation gain conversion model and wireless communication technology.

Accordingly, this creation provides real ear measurement and analysis without real ear analyzers, probe transducers (that is, probe microphones (probe microphone)), without being limited to a professional listening space, and without hearing professionals. assistance to effectively solve the above problems, and can be The current real environment in the non-audiometry room uses hearing aids to measure the real ear through wireless communication technology, and provides accurate, real-time, automatic and customized hearing aids for users.

1: Self-adjusting auditory compensation device with real ear measurement

11: First transducer

12: The first hearing compensation module

13: Second transducer

14: Wireless transmission and reception module

15: The third transducer

16: Storage module

10: Device

102: Second hearing compensation module

110: Equipment or devices

21: window frame

22: Discrete Fourier Transform (DFT)

23:|‧| ²

24: Logarithm (log(‧))

31: Hearing threshold A _f

32: Compensation Prescription

33: Compensation gain conversion model

34: Headphone device

與T _f 之間的誤差(

) 35: Calculate

The error between T f and T _f (

)

36: Judging whether the error target is met

超過A _f 的百分比)參數 37: Auditory dynamic range application optimization (

Percentage over A _f ) parameter

S1-S6: steps

S11-S17: Steps

S21-S27: Steps

FIG. 1 is a schematic block diagram of the self-adjusting auditory compensation device with real ear measurement of the present invention.

FIG. 2 is a schematic diagram of an embodiment of the self-adjusting auditory compensation device with real ear measurement combined with a smart device according to the present invention.

FIG. 3 is a flow chart of steps of the hearing compensation module of the self-adjusting hearing compensation device with real ear measurement through compensation gain conversion model technology and active noise cancellation (ANC) technology.

FIG. 4 is a schematic diagram showing model training of a compensation gain conversion model according to an embodiment of the present invention.

FIG. 5 shows a method for extracting a log-power spectrum (LPS) according to an embodiment of the present invention.

FIG. 6 is a block diagram illustrating model training of a compensation gain conversion model according to an embodiment of the present invention.

後的步驟流程圖。 Figure 7A is the application program side of this creation receiving electrical signals

Flowchart of subsequent steps.

7B is a schematic diagram of the target gain, hearing threshold and energy distribution of the self-adjusting auditory compensation device with real ear measurement of the present invention.

FIG. 8 is a flow chart of the steps of the self-adjusting auditory compensation method with real ear measurement of the present invention.

The implementation of this creation is described below through specific specific examples, and those who are familiar with the art can easily understand other advantages and effects of this creation from the content disclosed in this specification.

It should be noted that the structures, proportions, sizes, etc. shown in the drawings attached to this manual are only used to match the content disclosed in the manual, for the understanding and reading of those familiar with this technology, and are not used to limit the implementation of this creation Therefore, it has no technical substantive meaning. Any modification of structure, change of proportional relationship or adjustment of size shall still fall within the scope of this creation without affecting the effect and purpose of this creation. The technical content revealed by the creation must be within the scope covered.

，且透過無線傳送接收模組14及無線傳輸網路(未顯示於圖式中)傳送電性訊號

至該裝置(未顯示於圖式中)，其中，該裝置係利用應用程式(app)、其韌體或雲端技術計算電性訊號在各頻帶下的能量分佈(energy distribution)，且透過第二聽力補償模組102比較能量分佈與目標增益及聽力閾值的誤差，其中，若該誤差未符合誤差目標，則該裝置利用應用程式、其韌體或雲端技術對該誤差進行量化，以藉由補償增益轉換模型產生一組經修正之濾波器參數後，透過無線傳輸網路傳送該組經修正之濾波器參數至第一聽力補償模組12、第二聽力補償模組或其他具有音源處理能力(或聽力補償模組)的設備或裝置，以進行聽力增益補償。 FIG. 1 is a schematic block diagram of the self-adjusting auditory compensation device with real ear measurement of the present invention. According to the embodiment of the invention, as shown in Figure 1, the self-adjusting hearing compensation device 1 with real ear measurement of the invention includes a first transducer (Ref. Mic) 11, a first hearing compensation module 12, a second transducer Transducer (that is, loudspeaker) 13, wireless transmission and reception module 14, the third transducer (Err.Mic) 15 and storage module 16, wherein, the first transducer (Ref.Mic) 11 is received from a device (such as a smart device or mobile device) test signal S, and convert the test signal S into an electrical signal; the first hearing compensation module 12 is connected to the first transducer (Ref.Mic) 11 and the wireless Transmitting and receiving module 14, and performing gain compensation on the electrical signal; the second transducer 13 is connected to the first hearing compensation module 12, and converts the gain-compensated electrical signal into sound, and the sound transmitted to the ear canal; the wireless transmitting and receiving module 14 is connected to the first hearing compensation module 12; and the third transducer (Err.Mic) 15 is connected to the wireless transmitting and receiving module 14, the third transducer Transmitter 15 synchronously converts the sound transmitted in the ear canal into an electrical signal

, and transmit electrical signals through the wireless transmission and reception module 14 and the wireless transmission network (not shown in the figure)

To the device (not shown in the figure), wherein the device uses the application program (app), its firmware or cloud technology to calculate the energy distribution of the electrical signal in each frequency band (energy distribution), and through the second The hearing compensation module 102 compares the error of the energy distribution with the target gain and the hearing threshold, wherein, if the error does not meet the error target, the device uses the application program, its firmware or cloud technology to quantify the error, so as to compensate After the gain conversion model generates a set of modified filter parameters, the modified set of filter parameters is sent to the first hearing compensation module 12, the second hearing compensation module or other audio source processing capabilities ( or hearing compensation module) equipment or device for hearing gain compensation.

In an embodiment of the present invention, the first hearing compensation module 12 is set in an active noise reduction chip or a chip of a digital signal processing circuit, and the second hearing compensation module is set in the device (such as a smart phone) type device or mobile device), it is realized by application program, its firmware or cloud technology, wherein the first hearing compensation module 12 is synchronized with the second hearing compensation module.

In an embodiment of the present invention, the set of modified filter parameters are the filter parameters of gain compensation of ANC or the gain compensation parameters of digital signal processing circuit, wherein, the filter parameters of gain compensation of ANC Parameters are audio gain compensation filter unit (eg SZ or APT filter) parameters.

As shown in FIG. 1 , the self-adjusting auditory compensation device 1 with real ear measurement of the present invention further includes a storage module 16, wherein, if the above-mentioned error meets the error target, the device uses The application program, its firmware or cloud technology stores the set of modified filter parameters to the storage module 16 . Further, the device stores the original filter parameters in a device or device capable of processing audio sources, wherein the device or device has a hearing compensation module for hearing gain compensation. In one embodiment, the device or device with audio processing capability stores the original filter parameters in an active noise reduction chip or a chip with a digital signal processing circuit for hearing gain compensation.

In addition, in an embodiment of the present invention, if the error still does not meet the error target, the device uses the application program, its firmware or cloud technology to quantize the error again, so as to generate another set of errors by compensating the gain conversion model After correcting the filter parameters, send another set of corrected filter parameters to the first hearing compensation module, the second hearing compensation module or other devices with sound source processing capabilities (or hearing compensation modules) through the wireless transmission network Or a device for hearing gain compensation, wherein the compensation gain conversion model can be set in the cloud, server or smart device, the invention is not limited thereto.

According to another embodiment of the invention, as shown in Figure 1 and Figure 2, the wireless transmission and reception module of the self-adjusting auditory compensation device 1 with real ear measurement of the invention can also pass through the wireless transmission network (not shown in the figure) where) receives a test signal S from a device (such as a smart device or a mobile device) 10 . Similar to the above-mentioned embodiment, if the wireless transmission and reception module of the self-adjusting hearing compensation device 1 with real ear measurement receives the test signal S from the device 10 through a wireless transmission network (not shown in the drawings), then the first The hearing compensation module 12 performs gain compensation on the test signal S; the second transducer 13 is connected to the first hearing compensation module 12, converts the gain-compensated test signal into sound, and transmits the sound to the ear Inside the channel; the wireless transmission and reception module 14 is connected to the first hearing compensation module 12; the third transducer (Err.Mic) 15 is connected to the wireless transmission The receiving module 14, the third transducer 15 synchronously converts the sound transmitted in the ear canal into an electrical signal, and transmits the electrical signal to the device 10 through the wireless transmitting and receiving module 14 and the wireless transmission network, wherein , the device 10 uses an application program (app), its firmware or cloud technology to calculate the energy distribution of the electrical signal in each frequency band (energy distribution), and compares the energy distribution with the target gain and the target gain through the second hearing compensation module 102 The error of the hearing threshold, if the error does not meet the error target, the device 10 uses the application program, its firmware or cloud technology to quantify the error, so as to generate a set of modified filter parameters by compensating the gain conversion model , sending the set of modified filter parameters to the first hearing compensation module 12, the second hearing compensation module 102 or other equipment or devices with sound source processing capabilities (or hearing compensation modules) through the wireless transmission network, so as to Perform hearing gain compensation.

In another embodiment of the present invention, the first hearing compensation module 12 is set in an active noise reduction chip or a chip of a digital signal processing circuit, and the second hearing compensation module 102 is set in the device 10 (such as a smart device or a mobile device), it is realized by an application program, its firmware or cloud technology, wherein the first hearing compensation module 12 is synchronized with the second hearing compensation module 102 .

In an embodiment of the present invention, the set of modified filter parameters are the filter parameters of gain compensation of ANC or the gain compensation parameters of digital signal processing circuit, wherein, the filter parameters of gain compensation of ANC Parameters are audio gain compensation filter unit (eg SZ or APT filter) parameters.

However, in another embodiment of the present invention, if the error still does not meet the error target, the device uses the application program, its firmware or cloud technology to quantize the error again, so as to generate another error by compensating the gain conversion model After setting the corrected filter parameters, transmit them through the wireless transmission network Send another set of modified filter parameters to the first hearing compensation module, the second hearing compensation module or other equipment or devices with sound source processing capabilities (or hearing compensation modules) for hearing gain compensation, wherein the compensation The gain conversion model can be set in the cloud, server or smart device, and the invention is not limited thereto.

In the embodiment of the invention, the self-adjusting auditory compensation device with real ear measurement of the invention is set in a hearing aid with active noise reduction or digital signal processing circuit.

In addition, the above-mentioned modules can be hardware or firmware; if it is hardware, it can respectively realize various circuits of hearing gain compensation, wireless transmission and reception, and storage or a hardware unit with similar technology; if it is firmware, then It can be various firmware units for performing hearing gain compensation, wireless transmission and reception, and storage respectively. In one embodiment, the hearing compensation module can be a hearing compensation circuit or a hearing compensation hardware/firmware unit, the wireless transmission and reception module can be a wireless transmission and reception circuit or a wireless transmission and reception hardware/firmware unit, and the storage module can be For storing circuits or storing hardware/firmware units, the self-adjusting auditory compensation device of the present invention includes but is not limited to ANC.

The self-adjusting auditory compensation device with real ear measurement of this creation is set in the hearing aid without using an additional probe transducer (that is, probe microphone (probe microphone)), so that the current reality in the non-audiometry room The environment provides accurate, real-time, automatic and customized hearing aids through wireless communication technology.

，且透過無線傳送接收模組及無線傳輸網路(未顯示於圖式中)傳送電性訊號

至該裝置(未顯示於圖式中)，其中，該裝置計算電性訊號在各頻帶下的能量分佈(energy distribution)，且透過該裝置中之聽力補償模組比較能量分佈與目標增益及聽力閾值的誤差，若該誤差未符合誤差目標，則該裝置對該誤差進行量化，以藉由補償增益轉換模型產生一組經修正之濾波器參數後，透過無線傳輸網路傳送該組經修正之濾波器參數至聽力補償模組或其他具有音源處理能力(或聽力補償模組)的設備或裝置，以進行再一次聽力增益補償。 Specifically, this creation mainly combines application technology and compensation gain conversion model technology through noise reduction (for example, ANC) technology. In the current real environment of the user, there is no need to use additional probe transducers. This creation has The self-adjusting hearing compensation device for real ear measurement has included a wireless transmission and receiving module, a hearing compensation module, a transducer (that is, a loudspeaker), a transducer (Err.Mic), a transducer (Ref.Mic) and A storage module, wherein the transducer (Ref.Mic) receives a test signal S from a device and converts the test signal S into an electrical signal; the hearing compensation module is connected to the transducer (Ref.Mic) And the wireless transmission and reception module, and gain compensation for the electrical signal; the transducer is connected to the hearing compensation module, converts the gain-compensated electrical signal into sound, and transmits the sound to the ear canal ; and the transducer (Err.Mic) is connected to the wireless transmitting and receiving module, and the transducer simultaneously converts the sound transmitted in the ear canal into an electrical signal

, and transmit electrical signals through wireless transmission and reception modules and wireless transmission network (not shown in the figure)

To the device (not shown in the drawing), wherein the device calculates the energy distribution of the electrical signal in each frequency band, and compares the energy distribution with the target gain and hearing through the hearing compensation module in the device The error of the threshold value, if the error does not meet the error target, the device quantifies the error to generate a set of modified filter parameters by compensating the gain conversion model, and then transmits the modified set of filter parameters through the wireless transmission network The filter parameters are sent to the hearing compensation module or other equipment or devices with audio source processing capability (or hearing compensation module) to perform another hearing gain compensation.

FIG. 3 is a flow chart of steps of the hearing compensation module of the self-adjusting hearing compensation device with real ear measurement through compensation gain conversion model technology and active noise cancellation (ANC) technology.

First, in step S1, the application program (app) transmits the test signal S through the speaker of the smart device (or through the wireless transmission network), and then transmits the test signal S through the transducer (Ref.Mic) of the ANC headset (or wireless transmission The receiving module) receives the test signal S.

Then, in step S2, the filter circuit (or digital signal processing (DSP) circuit) of the ANC earphone performs hearing gain compensation through the hearing compensation module, and the transducer (ie, the speaker) in the ANC earphone to broadcast.

，且透過無線傳輸網路將該電性訊號

傳回智慧型裝置之應用程式端。 Afterwards, in step S3, the ANC earphone end synchronously uses the transducer (Err. Mic) to convert the sound signal in the ear canal into an electrical signal

, and the electrical signal through the wireless transmission network

Return to the application terminal of the smart device.

在各頻帶下的能量分佈(energy distribution)、目標增益及聽力閾值之特性，也就是，透過聽力補償模組比較該電性訊號

之能量分佈與目標增益及聽力閾值的誤差。 In step S4, the application side will simultaneously consider the calculated electrical signal

The characteristics of energy distribution, target gain and hearing threshold in each frequency band, that is, comparing the electrical signal through the hearing compensation module

The error of the energy distribution and target gain and hearing threshold.

In step S5, the compensation gain conversion model automatically generates modified filter parameters.

Finally, in step S6, if the error does not meet the error target, the application side automatically quantizes the error to generate corrected filter parameters through the compensation gain conversion model, and then transmits the corrected filter parameters through the wireless transmission network The filter parameters to the hearing compensation module to perform hearing gain compensation again; if the error meets the error target, the application automatically stores the corrected filter parameters to the storage module in the ANC headset and/or the in smart devices. Further, in another embodiment, the application automatically stores the original filter parameters to a device or device with audio processing capability (110 shown in FIG. 2, such as a smart device, a mobile device, a speaker or a sound box) ), wherein the device or device has a hearing compensation module for hearing gain compensation.

It is worth mentioning that if the above error still does not meet the error target, the application program will quantify the error again to generate another set of modified filter parameters through the compensation gain conversion model, and then transmit another set of parameters through the wireless transmission network. A set of modified filter parameters is sent to the hearing compensation module for another hearing gain compensation.

的能量分佈特性、目標增益、聽力閾值及聽覺動態範圍應用優化，補償增益轉換模型會自動地透過模型訓練產生複數組(或n組)ANC濾波器參數，可以提供ANC耳機進行聽力增益補償。 In an embodiment of the present invention, FIG. 4 shows a schematic diagram of model training of a compensation gain conversion model. based on electrical signals

The energy distribution characteristics, target gain, hearing threshold and auditory dynamic range application optimization, the compensation gain conversion model will automatically generate complex groups (or n groups) of ANC filter parameters through model training, which can provide ANC headphones for hearing gain compensation.

之能量分佈的計算以及透過補償增益轉換模型架構進行補償參數計算詳細說明。 The following are for electrical signals

The calculation of the energy distribution and the calculation of the compensation parameters through the compensation gain conversion model framework are detailed.

之能量分佈的計算 electrical signal

Calculation of the energy distribution

先傳送到APP端進行聲學特徵提取，且透過對數功率頻譜(log-power spectrum,LPS)方法來計算在一段時間下電性訊號

中的能量分佈情況。當特徵進行擷取時，通過計算各重疊窗框(window frame)21的離散傅立葉轉換(discrete Fourier transform,DFT)22，對輸入訊號進行短時傅立葉轉換(short-time Fourier transform,STFT)，亦即透過公式(1)將語音訊號從時域轉為頻域，公式(1)如下所示：

FIG. 5 shows a log-power spectrum (LPS) extraction method. In the embodiment of this creation, the transducer (Err.Mic) will receive the electrical signal

First send it to the APP for acoustic feature extraction, and use the logarithmic power spectrum (log-power spectrum, LPS) method to calculate the electrical signal over a period of time

The energy distribution in . When the feature is extracted, by calculating the discrete Fourier transform (discrete Fourier transform, DFT) 22 of each overlapping window frame (window frame) 21, the short-time Fourier transform (short-time Fourier transform, STFT) is performed on the input signal, and also That is, the speech signal is converted from the time domain to the frequency domain through the formula (1), and the formula (1) is as follows:

k =0 , 1 , ... ,L -1

)在時域中第l個樣本，Y ^f (k)代表輸入訊號的頻譜，k是頻率索引(frequency index)，h(l)表示漢明窗函數，對數功率譜定義如公式(2)所示：Y ^l (k)=log|Y ^f (k)|² (2) Among them, Y ^t ( l ) represents the input signal (ie, electrical signal

) in the l- th sample in the time domain, Y ^f ( k ) represents the spectrum of the input signal, k is the frequency index (frequency index), h ( l ) represents the Hamming window function, and the logarithmic power spectrum is defined as in formula (2) Show: Y ^l ( k )=log| Y ^f ( k )| ² (2)

k =0 , 1 , ... ,K -1

的對數功率頻譜Y ^l (k)。 Among them, Y ^l ( k ) represents the logarithmic power spectrum of the input signal, as shown in formula (2), the spectrum Y ^f ( k ) of the input signal is |. | ² 23, and take the logarithm (log(.))24 to obtain the electrical signal

The logarithmic power spectrum of Y ^l ( k ).

的能量分佈特性。 At this time, the logarithmic power spectrum under n overlapping window frames is accumulated to obtain the electrical signal

energy distribution characteristics.

Compensation gain parameters through compensation gain conversion model framework

FIG. 6 shows a block diagram of model training for a compensation gain switching model. After the hearing threshold A _f 31 obtained by the hearing screening, the user first passes the compensation prescription 32 (for example, NAL-R, NAL-RP, DSL, NAL-NL1, NAL-NL2, Aescu HRL-1...etc. ) to calculate the hearing gain compensation G _f required by the user. Then, the compensation gain is converted into the filter parameter gain G' _N required by the circuit through the compensation gain conversion model 33 (for example, deep learning method, machine learning method, mathematical statistics method, etc.), and sent to the ANC earphone The device 34, the above process can be model trained through the compensation gain conversion model framework, and realized through the target loss function (cost function), as shown in the formula (3):

Among them, N indicates that the model will generate complex sets of filter parameters (obtain N sets), M indicates the number of samples for training the model, and i indicates the number of gain data in training.

When the compensation gain conversion model is used for model training, the error is backpropagated to update the model parameters, and the parameter weights are adjusted to find the best compensation gain, as shown in formula (4):

錄製後與目標語音訊號T _f 分佈進行計算35，以獲得兩者訊號之間的誤差

，其中，計算誤差的方法包含以下方法：minimum mean-square error、客觀性評估指標(例如，HASQI,HASPI,STOI,NCM,PESQ,...等)，本創作不以此為限。之後，透過判別來確定當前的誤差是否在可以接受之範圍內36。若誤差

是在可以接受的範圍內時，則表示選配完成；反之，將此誤差

及聽覺動態範圍應用優化參數

37再次傳送至補償增益轉換模型來重新產生另一組修正後補償增益

。值得注意的是，聽覺動態範圍應用優化參數

為

超過A _f 的百分比之參數。經過上述流程，可以重複以上流程使誤差

持續收斂至符合設定之需求，從而完成自動化的選配流程。 Then, the electrical signal in the ear canal is transmitted through the transducer of the ANC device

Calculate 35 with the target speech signal T _f distribution after recording to obtain the error between the two signals

, wherein, the method of calculating the error includes the following methods: minimum mean-square error, objective evaluation index (for example, HASQI, HASPI, STOI, NCM, PESQ, ... etc.), this creation is not limited to this. Afterwards, it is determined whether the current error is within an acceptable range 36 through judgment. If the error

When it is within the acceptable range, it means that the matching is completed; otherwise, the error

and auditory dynamic range application optimization parameters

37 is sent again to the compensation gain conversion model to regenerate another set of modified compensation gains

. It is worth noting that the auditory dynamic range application optimizes the parameter

for

Parameters that exceed the percentage of A _f . After the above process, the above process can be repeated to make the error

Continue to converge until it meets the set requirements, thus completing the automated matching process.

之能量分佈的計算，且透過補償增益轉換模型架構進行增益補償參數。 It is worth mentioning that the DSP circuit can also perform electrical signal processing through the above-mentioned process.

The calculation of the energy distribution, and the gain compensation parameters are performed through the compensation gain conversion model framework.

後的步驟流程圖，且圖7B為本創作之具有真耳測量的自調式聽覺補償裝置的目標增益、聽力閾值及能量分佈之示意圖。 Figure 7A is the application program side of this creation receiving electrical signals

The following steps are flow charts, and FIG. 7B is a schematic diagram of the target gain, hearing threshold and energy distribution of the self-adjusting auditory compensation device with real ear measurement of the present invention.

傳送至智慧型裝置之應用程式端。 As shown in Figure 7A, in step S11, the ANC earphone end transmits the electrical signal through the transducer (Err.Mic)

Send to the application side of the smart device.

(例如，約10秒鐘的測試語句)。 Next, in step S12, the application program side receives the electrical signal

(e.g. about 10 seconds of test statement).

的能量分佈。 Afterwards, in step S13, take n sound frames, perform Fourier transform on each sound frame, and accumulate the energy under n sound frames to obtain electrical signals

energy distribution.

的能量分佈與聽力閾值及目標增益...等之間的誤差，若該誤差未符合該誤差目標，則對該誤差再進行量化以及傳送量化後的誤差及聽覺動態範圍應用優化參數至補償增益轉換模型，並藉由補償增益轉換模型產生另一組修正後濾波器參數。 In step S14, calculate and compare the current electrical signal

The error between the energy distribution of the hearing threshold and the target gain... If the error does not meet the error target, the error is quantized again and the quantized error and auditory dynamic range are transmitted to apply optimization parameters to the compensation gain transformation model, and generate another set of modified filter parameters by compensating the gain transformation model.

In step S15, the amount of gain in each frequency band is adjusted.

In step S16, the modified ANC filter parameters (or gain compensation parameters of the DSP circuit) are generated through the compensation gain conversion model after model training.

Finally, in step S17, if the above-mentioned error meets the error target, the ANC filter parameters are written into the chip of the ANC earphone (that is, stored in the storage module of the ANC earphone) and/or in the smart device; or, Write the gain compensation parameters of the DSP circuit into the chip of the earphone with the DSP circuit.

It is worth mentioning that the self-adjusting auditory compensation device with real ear measurement in this creation uses active noise cancellation (active noise cancellation, ANC) technology, but in different embodiments, the same or similar noise reduction technology can be used Applicable, this creation is not limited to this. In this embodiment, the parameters of the filters (e.g. FF, FB, SZ, APT...etc.) can be set through the information of the "acoustic characteristics of the mechanism" and the "acoustic compensation prescription", that is, the mean squared error will be The (mean-square error, MSE) method sets the filter parameters in the ANC technology, and then enables the ANC technology to perform gain compensation capabilities of different frequencies for the sound source transmitted by the transducer. In one embodiment, the above-mentioned filters can be feed-forward (FF) filter, feedback (FB) filter, audio gain compensation filter, respectively. Filter unit (such as SZ filter, APT filter), where the feedforward (FF) filter can receive the electrical signal of the transducer (Ref.Mic) to eliminate external noise; the feedback (FB) filter can Receive the electrical signal of the transducer (Err.Mic) (that is, the transducer (Err.Mic) converts the noise in the ear canal into an electrical signal) to eliminate the noise in the ear canal; and the audio gain compensation filter unit (such as SZ filter and APT filter) receive a suitable target curve (target curve) to amplify the signals of each frequency band in the electrical signal. In addition, in the DSP circuit, the time domain (or frequency domain) gain amplification unit in the processing architecture can be adjusted, such as EQ, filter, wide dynamic range compression (wide dynamic range compression), adaptive dynamic range optimization ( adaptive dynamic range optimization), etc.

Since this creation is applicable to various smart devices, the self-adjusting auditory compensation device can be used in non-audiometry room environments (such as residential buildings, outdoors, cars, parks, etc.) without the assistance of hearing professionals Conduct audiometry (audiometry), that is to say, the self-adjusting auditory compensation device of this creation does not have to be limited to the hearing test and real ear measurement analysis in the audiometry room with real ear measurement instruments, so as to provide the current real environment in the non-audiometry room Automated, real-time, user-customized hearing aids, auditory aids, or devices with hearing aids.

In one embodiment, the hearing compensation device with real ear measurement of the present invention is set on hearing aids, such as earphones (including but not limited to moving coil, moving iron, piezoelectric, pneumatic, electrostatic, Wired transmission, wireless transmission earphones), hearing aids, anti-noise earphones, monitor earphones, smart glasses, wearable devices or a combination thereof. In another embodiment, the hearing aid with real ear measurement of the present invention is also a hearing device, having the above-mentioned hearing compensation device, wherein the hearing compensation device is set and connected to the hearing device.

In addition, the self-adjusting auditory compensation device with real ear measurement of the present invention combines compensation gain conversion model technology and wireless communication technology (such as Bluetooth (Bluetooth), Wi-Fi, near Near-field communication (near-field communication, NFC), ultra-wideband (ultra-wideband, UWB), IEEE 802.15.4 and other wireless communication technologies), directly customize the audiogram of users (especially those with hearing loss) in real time Or the audiometer operates synchronously with the noise reduction module and/or hearing compensation module installed in the same or a single chip, thereby providing users (especially hearing-impaired patients) with a comfortable listening experience in real time. In addition, according to the above-mentioned embodiments of the present invention, since the hearing-impaired person uses his or her own hearing equipment or device (for example, various smart devices or devices with ANC earphones or TWS earphones) can be used in various current real environments or real application environments (that is, Quiet or noisy environment) instead of hearing tests in the audiology room, so the hearing-impaired can choose to turn on or turn off the noise reduction module when performing the self-adjusting hearing compensation of this creation according to their own needs.

It is worth noting that the self-adjusting auditory compensation device with real ear measurement in this creation does not need to be limited to the hearing test and/or hearing gain compensation in the audiology room and does not require the assistance of hearing professionals. This creation also does not need to use The additional probe transducer can be automatically, instantly and customer-controlled only through its own device (such as hearing aids, hearing aids, earphones, etc.) and through smart devices combined with compensation gain conversion model technology and wireless communication technology. Hearing aids, hearing aids, or earphones with hearing aid functions for hearing-impaired patients.

FIG. 8 is a flow chart of the steps of the self-adjusting auditory compensation method with real ear measurement according to the present invention, together with the description of the above-mentioned embodiment, wherein the method flow includes at least the following steps S21 to S27.

In step S21, a first test signal from a device (such as a smart device or a mobile device) is received by the first transducer, and the first test signal is converted into a first electrical signal.

In step S22, gain compensation is performed on the first electrical signal through the first hearing compensation module connected to the first transducer and the wireless transmission and reception module.

In step S23, the gain-compensated first electrical signal is converted into sound by the second transducer connected to the first hearing compensation module, and the sound is transmitted into the ear canal.

In step S24, the sound transmitted in the ear canal is converted into a second electrical signal synchronously by the third transducer connected to the wireless transmitting and receiving module, so as to be transmitted through the wireless transmitting and receiving module and the wireless transmission network The second electrical signal is sent to the device.

In step S25, the device uses the application program, its firmware or cloud technology to calculate the energy distribution of the second electrical signal in each frequency band, and compares the energy distribution with the target gain and hearing through the second hearing compensation module threshold error.

In step S26, if the error does not meet the error target, the device uses the application program, its firmware or cloud technology to quantify the error to generate a set of modified filter parameters by compensating the gain conversion model, Send the set of modified filter parameters to the first hearing compensation module and the second hearing compensation module through the wireless transmission network for hearing gain compensation.

In step S27, if the error meets the error target, the device uses the application program, its firmware or cloud technology to store the set of modified filter parameters to the storage module.

In another embodiment, in addition to storing the set of modified filter parameters in the storage module, the device can also use the application program, its firmware or cloud technology to store the original filter parameters in a computer with audio processing capabilities. The device or device, wherein the device or device has a hearing compensation module for hearing gain compensation.

In addition, in addition to the first transducer receiving the first test signal from a device (such as a smart device or mobile device), the wireless transmitting and receiving module can also receive the second test signal from the device through the wireless transmission network. signal for hearing gain compensation as described above. In addition, the first test signal is transmitted in the air, and the second test signal is transmitted through wireless communication.

In the above method process, if the error still does not meet the error target, the device uses the application program, its firmware or cloud technology to quantize the error again and transmit the quantized error and the auditory dynamic range application optimization parameters to the compensation gain conversion model, so that after another set of modified filter parameters is generated by the compensation gain conversion model, the another set of modified filter parameters is sent to the first hearing compensation model through the wireless transmission network set, the second hearing compensation module to perform the hearing gain compensation.

In the above-mentioned method flow, the first hearing compensation module is set in an active noise reduction chip or a chip of a digital signal processing circuit, and the second hearing compensation module is set in the device (such as a smart device) or a mobile device), implemented by an application program, its firmware or cloud technology, wherein the first hearing compensation module is synchronized with the second hearing compensation module.

In one embodiment, the set of modified filter parameters are filter parameters for gain compensation of active noise reduction or gain compensation parameters of a digital signal processing circuit, wherein the filter parameters for gain compensation of active noise reduction are Audio gain compensation filter unit (eg SZ or APT filter) parameters.

In addition, the above-mentioned method flow is applied to a self-adjusting auditory compensation device, and can also be applied to a hearing aid with an active noise reduction or digital signal processing circuit.

It is worth noting that if the above error still does not meet the error target, the device will use the application program, its firmware or cloud technology to quantify the error again, so as to generate another set of modified filter by the compensation gain conversion model After the filter parameters are obtained, another set of modified filter parameters is sent to the hearing compensation module through the wireless transmission network to perform another hearing gain compensation.

To sum up, the self-adjusting auditory compensation device with real ear measurement and the self-adjusting auditory compensation method with real ear measurement in this creation are based on active noise reduction (ANC) technology combined with digital network technology and wireless transmission technology, not only It enables the earphones to emit a reverse wave (or forward wave) with the same energy as the current noise to eliminate the environmental noise in the ear canal, and it can also directly monitor the user through the hearing compensation module when performing real ear measurement (REM). (Especially for hearing-impaired patients) Real-time customized audiogram or audiometer for hearing gain compensation (and thus amplifying the signals of each frequency band (such as forward signal and/or reverse signal)), with automatic, real-time and customized Hearing aids, hearing aids, or earphones with hearing aid function for hearing-impaired patients.

In addition, the self-adjusting auditory compensation device with real ear measurement and the self-adjusting auditory compensation method with real ear measurement in this creation consider the hearing loss characteristics of the hearing-impaired person through the compensation gain conversion model technology to provide a representative test for the hearing-impaired person Words, and then perform real ear measurement, and achieve the effect of automatic, real-time and customized hearing aids, hearing aids or earphones with hearing aid functions for the hearing-impaired.

In addition, in the embodiment of this invention, the compensation gain conversion model can also automatically modify the compensation parameters of the self-adjusting auditory compensation device (for example, SII (speech intelligibility index), HASQI, HASPI and other compensation parameters), wherein, the compensation gain conversion model can be set in the cloud, server or smart device, and this creation is not limited thereto.

Finally, in the embodiment of this creation, a computer program product uses the application program of the device, its firmware or cloud technology to execute the above content, and can automatically store the original filter parameters to the device with audio processing capabilities or a device (110 shown in FIG. 2 , such as a smart device, a mobile device, a speaker or a sound box), wherein the device or device has a hearing compensation module for hearing gain compensation. Therefore, the computer program product can choose to synchronize the modified filter parameters to the self-adjusting auditory compensation device or synchronize the original filter parameters to the device or device capable of processing audio sources for audio source processing and playback.

The above-mentioned embodiments are only illustrative to illustrate the principles and effects of the invention, and are not intended to limit the invention. Any person who is familiar with this technology can modify and change the above-mentioned implementation form without violating the spirit and scope of this creation. Therefore, the scope of protection of the rights of this creation should be listed in the scope of the patent application.

1: Self-adjusting auditory compensation device with real ear measurement

11: First transducer

12: The first hearing compensation module

13: Second transducer

14: Wireless transmission and reception module

15: The third transducer

16: Storage module

Claims (10)

A self-adjusting auditory compensation device with real ear measurement, comprising: a first transducer, which receives a first test signal from a device, and converts the first test signal into a first electrical signal; a first hearing The compensation module is connected to the first transducer and the wireless transmission and reception module, and performs gain compensation on the first electrical signal; the second transducer is connected to the first hearing compensation module, and converting the gain-compensated first electrical signal into sound, and transmitting the sound into the ear canal; and a third transducer connected to the wireless transmitting and receiving module, the third transducer synchronously transmitting The sound transmitted in the ear canal is converted into a second electrical signal, so as to transmit the second electrical signal to the device through the wireless transmitting and receiving module and through the wireless transmission network, wherein the device calculates the second electrical signal The energy distribution of the sexual signal in each frequency band, and compare the error between the energy distribution and the target gain and hearing threshold through the second hearing compensation module. If the error does not meet the error target, the device quantizes the error to borrow After the modified filter parameters are generated from the compensation gain conversion model, the modified filter parameters are sent to the first hearing compensation module through the wireless transmission network for hearing gain compensation. The self-adjusting hearing compensation device with real ear measurement as described in Claim 1, wherein the first hearing compensation module is set in an active noise reduction chip or a chip of a digital signal processing circuit, and the second hearing compensation The module system is set in the device. The self-adjusting auditory compensation device with real ear measurement as described in Claim 1, wherein the modified filter parameter is a gain compensation filter parameter of active noise reduction or a gain compensation parameter of a digital signal processing circuit. The self-adjusting auditory compensation device with real ear measurement as described in Claim 3, wherein the filter parameter of the gain compensation of the active noise reduction is an audio gain compensation filter unit parameter. The self-adjusting auditory compensation device with real ear measurement as described in claim 1 further includes: a storage module, wherein, if the error meets the error target, the device stores the corrected filter parameters to the storage module Group. The self-adjusting auditory compensation device with real ear measurement as described in Claim 1, wherein the device stores the original filter parameters to a device with sound source processing capability, wherein the device has a third hearing compensation module to Perform hearing gain compensation. The self-adjusting hearing compensation device with real ear measurement as described in Claim 1, wherein the wireless transmitting and receiving module receives the second test signal from the device through the wireless transmission network to perform hearing gain compensation. The self-adjusting auditory compensation device with real ear measurement as described in Claim 1, wherein, if the error still does not meet the error target, the device quantizes the error again and transmits the quantized error and auditory dynamic range applying optimization parameters to the compensation gain conversion model to generate another modified filter parameter through the compensation gain conversion model, and then transmitting the other modified filter parameter to the first hearing compensation model through the wireless transmission network group for this hearing gain compensation. The self-adjusting hearing compensation device with real ear measurement as described in Claim 1, wherein the self-adjusting hearing compensation device is set in a hearing aid with active noise reduction or digital signal processing circuit. A hearing device with the self-adjusting hearing compensation device according to any one of Claims 1-9, wherein the self-adjusting hearing compensation device is set and connected to the hearing device.

Images