Method, device and hearing aid for hearing aid audiometry based on environmental sound analysis

Abstract

This invention relates to the field of medical device technology, and discloses a method, device, and hearing aid for hearing aid audiometry based on ambient sound analysis. The hearing aid audiometry method provided by this invention includes: acquiring the ambient sound of the current environment; determining the signal-to-noise ratio (SNR) corresponding to the ambient sound; adjusting the audiometry configuration of the hearing aid according to the SNR, and controlling the hearing aid to perform audiometry according to the audiometry configuration. This invention can automatically adjust the audiometry configuration of the hearing aid based on real-time sound field environment and noise conditions, enabling hearing tests for hearing aid users under different noise conditions, improving the accuracy and reliability of test results, adapting to the user's actual environment, providing personalized hearing support, and thus providing users with clearer and more accurate sound enhancement effects.

Description

Technical Field

[0001] This invention relates to the field of medical device technology, specifically to a method, device, and hearing aid for hearing aid audiometry based on environmental sound analysis. Background Technology

[0002] Currently, hearing aid users are often disturbed by ambient noise during hearing tests, which negatively impacts the accuracy of the results. Although current technology tends to conduct hearing tests in relatively quiet environments, this approach is not always suitable for the constantly changing noise conditions in daily life. Summary of the Invention

[0003] In view of this, the present invention provides a method, device and hearing aid for hearing aid audiometry based on environmental sound analysis, so as to solve the problem of low accuracy and reliability of hearing aids due to interference from environmental noise during the audiometry process.

[0004] In a first aspect, the present invention provides a hearing test method for a hearing aid, the method comprising:

[0005] Get the ambient sound of the current environment;

[0006] Determine the signal-to-noise ratio corresponding to the ambient sound;

[0007] Based on the signal-to-noise ratio, adjust the audiometry configuration of the hearing aid and control the hearing aid to perform audiometry according to the audiometry configuration.

[0008] In one optional implementation, before adjusting the audiometry configuration of the hearing aid based on the signal-to-noise ratio, the method further includes:

[0009] Obtain the sound spectrum corresponding to the ambient sound;

[0010] The environmental noise energy at key frequencies in the sound spectrum is obtained, where the key frequencies are predetermined hearing aid audiometry frequencies.

[0011] The step of adjusting the audiometry configuration of the hearing aid based on the signal-to-noise ratio includes:

[0012] The audiometry configuration of the hearing aid is adjusted based on the ambient noise energy and the signal-to-noise ratio.

[0013] In one optional implementation, acquiring the environmental noise energy at key frequencies in the sound spectrum includes:

[0014] The average power value of the power spectrum at each time point and the power value corresponding to the key frequency are obtained. The power spectrum at each time point is the power spectrum of each short frame of the ambient sound segmentation.

[0015] Based on the average power value, a noise discrimination threshold is set;

[0016] The power value corresponding to each of the key frequencies is compared with the noise discrimination threshold.

[0017] The key frequencies whose power values ​​exceed the noise discrimination threshold are identified as noise frequencies;

[0018] The environmental noise energy is determined based on the power value of the noise frequency.

[0019] In one optional implementation, adjusting the audiometry configuration of the hearing aid based on the ambient noise energy and the signal-to-noise ratio includes:

[0020] Obtain the relationship between the ambient noise energy and the preset noise energy threshold;

[0021] If, based on the aforementioned size relationship, it is determined that the current environment is suitable for audiometry, then the noise reduction depth in the audiometry configuration is adjusted according to the signal-to-noise ratio.

[0022] In one optional implementation, adjusting the audiometry configuration of the hearing aid based on the signal-to-noise ratio includes:

[0023] Obtain the preset signal-to-noise ratio threshold corresponding to the signal-to-noise ratio;

[0024] If the signal-to-noise ratio is higher than the preset signal-to-noise ratio threshold, then the noise reduction depth in the audiometry configuration is reduced;

[0025] If the signal-to-noise ratio is lower than the preset signal-to-noise ratio threshold, then the noise reduction depth in the audiometry configuration is increased.

[0026] In one optional implementation, after acquiring the environmental noise energy at key frequencies in the sound spectrum, the method further includes:

[0027] Based on the frequency and intensity corresponding to the ambient noise energy, the masking threshold of the target signal of the ambient sound in each frequency band is calculated; the target signal is the audio signal used for listening measurement.

[0028] The target signal of the ambient sound is compared with the corresponding masking threshold, and the audibility of the target signal of the ambient sound is corrected based on the comparison result.

[0029] In one optional implementation, obtaining the sound spectrum corresponding to the ambient sound includes:

[0030] The sound field environment corresponding to the ambient sound is obtained using a neural network model;

[0031] The sound spectrum corresponding to the sound field environment is obtained as the sound spectrum corresponding to the ambient sound.

[0032] In one optional implementation, obtaining the sound field environment corresponding to the ambient sound using a neural network model includes:

[0033] Extract the Mel-frequency cepstral coefficients of the ambient sound;

[0034] Based on the Mel frequency cepstral coefficients, a corresponding audio feature vector is generated;

[0035] The neural network model is used to classify and evaluate the audio feature vectors to obtain the corresponding sound field environment.

[0036] In one optional implementation, extracting the Mel-frequency cepstral coefficients of the ambient sound includes:

[0037] The ambient sound is segmented into multiple short frames;

[0038] A preset window function is applied to each of the aforementioned short frames;

[0039] Perform a fast Fourier transform on each short-time frame for which the preset window function is applied, and calculate the power spectrum of each short-time frame;

[0040] The power spectrum of each short frame is passed through a set of Mel filter banks to generate multiple filter outputs;

[0041] Take the logarithm of the energy output of any filter, and perform a discrete cosine transform on the logarithmic energy value to generate cepstral coefficients;

[0042] The first N cepstral coefficients are selected as the Mel frequency cepstral coefficients of the ambient sound.

[0043] In one alternative implementation, the hearing aid audiometry method further includes:

[0044] If there are multiple mixed sounds in the sound field environment, the direction of the test sound source is determined based on the directionality estimation algorithm, and the signal reception intensity in the corresponding direction is enhanced.

[0045] In a second aspect, the present invention provides a hearing test device for a hearing aid, the device comprising:

[0046] The ambient sound acquisition module is used to acquire the ambient sound of the current environment;

[0047] The signal-to-noise ratio determination module is used to determine the signal-to-noise ratio corresponding to the ambient sound.

[0048] The audiometry module is used to adjust the audiometry configuration of the hearing aid according to the signal-to-noise ratio, and to control the hearing aid to perform audiometry according to the audiometry configuration.

[0049] Thirdly, the present invention provides a hearing aid, comprising: a memory and a processor, wherein the memory and the processor are communicatively connected to each other, the memory stores computer instructions, and the processor executes the computer instructions to perform the hearing test method of the hearing aid described in the first aspect or any corresponding embodiment.

[0050] Fourthly, the present invention provides a computer-readable storage medium storing computer instructions for causing a computer to perform the hearing aid measurement method of the first aspect or any corresponding embodiment described above.

[0051] Fifthly, the present invention provides a computer program product, including computer instructions for causing a computer to execute the hearing aid testing method of the first aspect or any corresponding embodiment described above.

[0052] The hearing aid testing method, device, hearing aid, and storage medium provided in this embodiment acquire the ambient sound of the user's current environment, determine the signal-to-noise ratio (SNR) corresponding to the ambient sound, and finally adjust the hearing aid's testing configuration according to the SNR. The hearing aid is then controlled to perform hearing tests according to the adjusted configuration. In other words, the hearing aid can automatically adjust its testing configuration based on the real-time sound field environment and noise conditions, enabling hearing tests for users under different noise conditions. This improves the accuracy and reliability of the test results, adapts to the user's actual environment, provides personalized hearing support, and ultimately offers clearer and more accurate sound enhancement. Attached Figure Description

[0053] To more clearly illustrate the technical solutions in the specific embodiments or related technologies of the present invention, the drawings used in the description of the specific embodiments or related technologies will be briefly introduced below. Obviously, the drawings described below are some embodiments of the present invention. For those skilled in the art, other drawings can be obtained from these drawings without creative effort.

[0054] Figure 1 This is a schematic flowchart of a hearing aid audiometry method according to an embodiment of the present invention;

[0055] Figure 2 This is a structural block diagram of the audiometry device for a hearing aid according to an embodiment of the present invention;

[0056] Figure 3 This is a schematic diagram of the hardware structure of a hearing aid according to an embodiment of the present invention. Detailed Implementation

[0057] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0058] According to an embodiment of the present invention, an embodiment of a hearing aid audiometry method is provided. It should be noted that the steps shown in the flowchart in the accompanying drawings can be executed in a computer system such as a set of executable computer instructions. Furthermore, although a logical order is shown in the flowchart, in some cases, the steps shown or described may be executed in a different order than that shown here.

[0059] This embodiment provides a hearing test method for hearing aids, which can be used in the hearing test device within the hearing aid, specifically various terminals, servers, etc. Figure 1 This is a flowchart of a hearing aid testing method according to an embodiment of the present invention, such as... Figure 1 As shown, the process includes the following steps:

[0060] Step S101: Obtain the ambient sound of the current environment. The current environment is the required test environment.

[0061] Specifically, ambient sounds of the current environment can be captured by the hearing aid. The hearing aid has a built-in microphone to capture sound signals from the surrounding environment; this is the first step in realizing the hearing aid's core function (amplifying and compensating for hearing loss). The hearing aid works as follows: the microphone converts external sounds (sound waves) into electrical signals; the captured electrical signals are then analyzed and processed by a digital signal processor (DSP), including noise suppression, frequency compensation, and gain adjustment; finally, the processed signal is amplified and converted into sound, which is transmitted to the user's ear canal through the receiver. Ambient sounds can include various background noises, such as traffic noise on the street, conversations among people indoors, and the operation of electronic devices in an office.

[0062] In some optional embodiments, before step S102, i.e. before adjusting the audiometry configuration of the hearing aid according to the signal-to-noise ratio, the method further includes:

[0063] Step 1: Obtain the sound spectrum corresponding to the ambient sound. The sound spectrum may include the sound energy distribution across various frequency bands from low to high frequencies.

[0064] In some optional implementations, step one, namely obtaining the sound spectrum corresponding to the ambient sound, includes:

[0065] Step 1 (a): Use a neural network model to obtain the sound field environment corresponding to the ambient sound.

[0066] In some optional implementations, step one (a), namely, using a neural network model to obtain the sound field environment corresponding to the ambient sound, includes:

[0067] Step 1 (a1): Extract the Mel-frequency cepstral coefficients of the ambient sound. Before extracting the Mel-frequency cepstral coefficients, the ambient sound can be preprocessed. Preprocessing may include, for example, noise reduction to reduce the impact of ambient noise on feature extraction. Alternatively, it may include pre-emphasis: applying a high-pass filter to enhance the high-frequency components, thereby increasing the energy of the high-frequency components and reducing the impact of high-frequency loss during signal transmission.

[0068] Specifically, step one (a1), which involves extracting the Mel-frequency cepstral coefficients of the ambient sound, includes:

[0069] Step 1 (a11): The ambient sound is segmented into multiple short-time frames.

[0070] In this context, a short-time frame refers to the process of dividing a continuous signal into a series of short time windows, where the signal within each window can be approximated as stationary. For speech signals, a typical short-time frame length is 20-40 milliseconds, which roughly corresponds to the duration of a syllable or formant period in human voice. Furthermore, setting the short-time frame length to 20-40 milliseconds ensures that sufficient audio features are captured during analysis while avoiding latency caused by excessively long frames. The frame shift during short-time frame segmentation can be around 10 milliseconds, with overlap between adjacent short-time frames.

[0071] Step one (a12) applies a preset window function to each of the short-time frames. The choice of window function is crucial for reducing spectral leakage; commonly used window functions include the Hamming window, Hanning window, and Blackman window.

[0072] Step one (a13) involves performing a Fast Fourier Transform (FFT) on each short-time frame for which a preset window function is applied, and calculating the power spectrum of each short-time frame. The Fast Fourier Transform (FFT) is a crucial step in converting a time-domain signal into a frequency-domain signal, providing frequency component information for each short-time frame. In other words, by converting the time-domain signal into a frequency-domain representation through the Short-Time Fourier Transform (STFT), the spectral information of the short-time frame is obtained, facilitating the analysis of the frequency composition of the ambient sound signal.

[0073] Step one (a14) involves passing the power spectrum of each short frame through a set of Mel filter banks to generate multiple filter outputs. This Mel filter bank can be, for example, a triangular filter bank distributed according to the Mel scale. These filter outputs reflect the perceptual characteristics of the human ear for different frequencies of sound. That is, through the processing of the Mel filter bank, a spectral representation closer to human perception can be obtained, because the Mel scale is designed based on the nonlinear characteristics of human ear perception of different frequencies of sound. In this embodiment of the invention, given the different sensitivities of the human ear to different frequencies of sound, the linear frequency scale is converted into a nonlinear Mel scale. This process is simulated using a Mel filter bank composed of a series of triangular filters. Each filter covers a frequency range, with denser filters in the low-frequency region and sparser filters in the high-frequency region. The sum of the energy outputs of each filter is calculated to obtain a set of energy values.

[0074] Step one (a15) involves taking the logarithm of the energy output of any filter and performing a Discrete Cosine Transform (DCT) on the logarithmic energy value to generate cepstral coefficients. The cepstral coefficients obtained through DCT further compress the audio feature data while retaining information useful for classification and evaluation. Here, taking the logarithm of the filter bank energy obtained in the previous step aims to compress the dynamic range, as the human ear perceives changes in sound intensity non-linearly.

[0075] Step one (a16) involves selecting the first N cepstral coefficients as the Mel-frequency cepstral coefficients of the ambient sound. Selecting the first N cepstral coefficients as the feature vector effectively reduces computation while maintaining sufficient classification accuracy. The value of N can be adjusted according to different computational resources and accuracy requirements. For example, on resource-constrained devices, a smaller value of N can be chosen to reduce computational burden; while in situations requiring higher accuracy, the value of N can be increased to improve classification accuracy.

[0076] In this embodiment of the invention, a Discrete Cosine Transform (DCT) is applied to the logarithmic energy of the filter bank to obtain the MFCC coefficients. The DCT transforms data from the spatial domain to the frequency domain, similar to the Fourier Transform, but it only uses cosine functions as basis functions. Typically, only the first N (e.g., 13) DCT coefficients are retained as the final MFCC eigenvectors because these low-order coefficients contain most of the information.

[0077] Step one (a2): Based on the Mel-frequency cepstral coefficients (MFCCs), generate the corresponding audio feature vector. MFCCs are feature parameters widely used in speech recognition and audio signal processing. By using MFCCs, the features of ambient sound can be effectively captured, thereby improving the accuracy of neural networks in classifying sound field environments. MFCCs are a form of audio feature vector.

[0078] Step one (a3): Using the neural network model, the audio feature vectors are classified and evaluated to obtain the corresponding sound field environment. Specifically, the audio feature vectors can be input into a pre-trained neural network model for classification and evaluation. By learning a large number of audio feature vectors under different sound field environments, the neural network model can accurately identify the sound field environment corresponding to the current ambient sound, thereby providing accurate environmental information for the hearing aid's audiometry configuration. The sound field environment can include indoor and outdoor environments, such as conference rooms, studies, libraries, tea rooms, restaurants, bustling streets, parks, and streets.

[0079] In this embodiment of the invention, a neural network model is used to analyze and obtain the sound field environment corresponding to ambient sounds. Specifically, the neural network calculates the probability value of each scenario based on the audio feature vector corresponding to the ambient sounds, and selects the scenario with the highest probability value for output, thereby determining the sound field environment in which the user is currently located. For example, on a noisy street, the hearing aid will collect continuous traffic noise. Then, the neural network will calculate the probability value of each scenario based on the sound characteristics (i.e., audio feature vector) of the traffic noise. For example, the probability of a street is 80%, the probability of a park is 30%, the probability of a restaurant is 63%, and the probability of a library is 14%, and finally determine that the user is in a street environment. In this embodiment of the invention, the sound field environment can be accurately classified, providing necessary environmental information for the hearing aid's audiometry configuration, thereby optimizing the user's auditory experience.

[0080] In this embodiment of the invention, the sound field environment is determined first, and then the corresponding sound spectrum is obtained. This improves the accuracy of subsequent ambient sound analysis, and consequently, the accuracy of audiometry configuration adjustments. This is because the characteristics of the sound field environment directly affect phenomena such as the propagation, reflection, and absorption of sound signals, thus affecting the final received sound spectrum. Directly obtaining the sound spectrum corresponding to the ambient sound without considering the sound field environment may lead to misjudgment of sound characteristics or inaccurate analysis results.

[0081] Step 1(b): Obtain the sound spectrum corresponding to the sound field environment as the sound spectrum corresponding to the ambient sound.

[0082] The sound spectrum can be obtained by performing frequency domain analysis on the sound signal; specifically, this involves converting the time-domain signal into a frequency-domain signal. The following are the detailed steps and methods for obtaining the sound spectrum:

[0083] 1. Digitize the sound signal (such as the ambient sound mentioned above); the digitized signal is a discrete time series x[n], which represents the amplitude of the sound at each sampling point.

[0084] 2. For non-stationary signals (such as speech), it is necessary to segment them into short-time frames.

[0085] 3. After framing, each frame of signal is multiplied by a window function (such as a Hanning window or a rectangular window) to reduce spectral leakage. The function of the window function is to smooth the boundaries of the signal and avoid artifacts caused by abrupt changes.

[0086] 4. Apply Discrete Fourier Transform (DFT) or Fast Fourier Transform (FFT) to the windowed signal.

[0087] 5. Calculate the spectral amplitude. To better visualize the spectrum, the amplitude can be converted to logarithmic form.

[0088] Step 2: Obtain the environmental noise energy at key frequencies in the sound spectrum. These key frequencies are pre-determined hearing aid audiometry frequencies. These key frequencies include 250, 500, 1000, 2000, 4000, and 6000 rpm, all of which are hearing aid audiometry points. These key frequencies are of particular importance in the use of hearing aids because they are typically associated with key components of the speech signal.

[0089] In some optional implementations, step two, namely obtaining the environmental noise energy of key frequencies in the sound spectrum, includes:

[0090] Step 2(a) involves obtaining the average power value of the power spectrum at each time point and the power value corresponding to the key frequency. The power spectrum at each time point is the power spectrum of each short frame of the ambient sound segmentation. For the power spectrum of each short frame of the ambient sound segmentation, please refer to the short frame power spectrum involved in the process of extracting the Mel-frequency cepstral coefficients of the ambient sound described above. The average power value of the power spectrum is obtained by summing the power values ​​at each frequency point and dividing by the number of frequency points. By calculating the average power value, an index reflecting the overall ambient noise level can be obtained.

[0091] The "time point" here actually refers to the time position of a short frame. The ambient sound is divided into multiple short frames, each representing an audio signal within a specific time period. After applying a window function and performing a Fast Fourier Transform (FFT) on each short frame, its power spectrum can be obtained.

[0092] Therefore, a "time point" can be understood as a time marker at the center or beginning of a short frame. For example, if the short frame length is 25 milliseconds and the frame shift is 10 milliseconds, then the first short frame covers a time range of [0, 25] milliseconds, and its corresponding time point could be 12.5 milliseconds (the center position). The second short frame covers a time range of [10, 35] milliseconds, and its corresponding time point is 22.5 milliseconds, and so on.

[0093] The sound spectrum generally refers to the result obtained after a signal undergoes a Fourier transform, showing the amplitude distribution of the signal at various frequencies. The power spectrum, on the other hand, further represents the power distribution of the signal at different frequencies, i.e., the signal power per unit frequency band. In other words, the power spectrum retains the amplitude information from the spectrum but loses the phase information.

[0094] From a mathematical perspective, the power spectrum can be viewed as the squared magnitude of each frequency component in the sound spectrum. For discrete signals, the power spectrum can be obtained by taking the modulus of the complex result after the Fast Fourier Transform (FFT) and squaring it.

[0095] Power value typically refers to the energy transmitted per unit time, while in signal processing, it can be understood as the average energy consumed per second by a signal at a certain frequency or across an entire frequency band. In contrast, energy value refers to the total energy accumulated over a period of time.

[0096] Step 2(b): Based on the average power value, set a noise discrimination threshold. The noise discrimination threshold can be, for example, a preset multiple of the average power value.

[0097] Step 2(c): Compare the power value corresponding to each of the key frequencies with the noise discrimination threshold to determine the environmental noise energy.

[0098] Key frequency points refer to the audiometry frequencies used by hearing aids. These frequencies are particularly important in the use of hearing aids because they are typically associated with key components of the speech signal. Determining the ambient noise energy at these key frequency points helps distinguish which frequency components are significantly affected by noise in the current sound field. By setting a noise discrimination threshold, frequency points with power values ​​exceeding the threshold can be identified as noise frequencies, thus providing a basis for subsequent signal-to-noise ratio calculations.

[0099] Specifically, the process for determining environmental noise energy is as follows: the power value of each key frequency is compared with a noise discrimination threshold, and then the key frequencies whose power values ​​exceed the noise discrimination threshold are identified as noise frequencies; finally, the environmental noise energy is determined based on the power value of the noise frequencies. For example, the average power value of multiple noise frequencies can be determined as the environmental noise energy.

[0100] In this embodiment of the invention, starting from key frequency points, noise frequency points are screened using a noise discrimination threshold, and finally, the environmental noise energy is derived by comprehensively considering the energy values ​​of these noise frequency points. This method ensures an accurate assessment of the impact of noise on the sound field environment.

[0101] Step S102: Determine the signal-to-noise ratio corresponding to the ambient sound.

[0102] Specifically, the signal-to-noise ratio (SNR) is the ratio of ambient noise energy to signal energy, reflecting the clarity of a signal at a specific frequency. SNR calculation is crucial for hearing aid performance because it directly affects whether the user can clearly hear the desired sound. After determining the SNR, the hearing aid's audiometric configuration will be adjusted accordingly to optimize the user's auditory experience. For example, if the SNR is low, the hearing aid may increase the gain at specific frequency points to improve the user's perception of those frequencies.

[0103] The signal-to-noise ratio (SNR) is calculated based on the relationship between environmental noise energy and signal energy. The SNR calculation process involves multiple steps, including environmental sound acquisition, sound field environment analysis, determination of environmental noise energy at key frequencies, and finally, calculation of the SNR based on these data.

[0104] Ambient sound acquisition and sound field environment analysis:

[0105] First, in step S100, the hearing aid acquires ambient sounds from the user's surroundings via its built-in microphone. These ambient sounds may include various background noises, such as traffic noise on the street or conversations among people indoors. Next, in step S200, a neural network is used to analyze these ambient sounds to determine their corresponding sound field environment, and the sound spectrum corresponding to this sound field environment is further analyzed to determine the environmental noise energy at key frequencies in the sound spectrum.

[0106] Determination of ambient noise energy at key frequencies:

[0107] The sound spectrum encompasses the distribution of sound energy across various frequency bands, from low to high frequencies. By analyzing this sound spectrum data, we can identify which frequency bands have higher sound energy, thus determining key frequencies. By analyzing the sound spectrum of ambient sound, we can calculate the noise energy value at each key frequency point. For example, for the 250Hz frequency point, a specific algorithm can be used to calculate the noise energy value at that frequency point, and then this process can be repeated for all key frequency points.

[0108] Signal-to-noise ratio calculation:

[0109] Once the ambient noise energy at the key frequency points is determined, the signal-to-noise ratio (SNR) can be calculated. The SNR is defined as the ratio of signal power to noise power, usually expressed in decibels (dB). The formula is as follows:

[0110]

[0111] Where Ps represents the average power of the signal, and Pn represents the average power of the noise. In this scenario, the signal power Ps can be regarded as the energy of the sound (such as speech) that the user expects to hear, while the noise power Pn is the environmental noise energy at the key frequency points calculated from the above steps.

[0112] Step S103: Adjust the audiometric configuration of the hearing aid according to the signal-to-noise ratio, and control the hearing aid to perform audiometry according to the audiometric configuration. The audiometric configuration may include parameters such as gain settings, frequency response adjustments, noise suppression levels, noise reduction depth, and sound amplification strategies. These parameters are adjusted based on the user's specific hearing needs and environmental noise conditions. For example, for users with high-frequency hearing loss, in noisy environments, the hearing aid can increase the gain in the high-frequency band while decreasing the gain in the low-frequency band to reduce interference from low-frequency noise. Furthermore, the hearing aid can dynamically adjust these parameters according to changes in environmental noise to provide the best auditory experience. For example, in a quiet library environment, the hearing aid can reduce gain to avoid amplifying unnecessary background noise; while in a noisy street, the hearing aid can increase gain to ensure the user can hear important sound signals. This provides users with more personalized and adaptive hearing support.

[0113] Step S103, namely adjusting the audiometry configuration of the hearing aid according to the signal-to-noise ratio, includes:

[0114] Step S1031: Adjust the audiometry configuration of the hearing aid based on the ambient noise energy and the signal-to-noise ratio. The process for acquiring the ambient noise energy is described above.

[0115] In some optional embodiments, step S1031, namely adjusting the audiometry configuration of the hearing aid based on the ambient noise energy and the signal-to-noise ratio, includes:

[0116] Step S10311: Obtain the relationship between the environmental noise energy and the preset noise energy threshold.

[0117] Step S10312: If, based on the size relationship, it is determined that the current environment is suitable for audio measurement, then the noise reduction depth in the audio measurement configuration is adjusted according to the signal-to-noise ratio.

[0118] In this embodiment, the current sound field environment is first compared with a preset noise energy threshold to determine whether it is suitable for audiometry. If the ambient noise energy is lower than the preset threshold, it indicates that the current ambient noise level is low and suitable for audiometry. Conversely, if the ambient noise energy is higher than the preset threshold, it may be necessary to take measures to reduce the noise impact or choose to conduct the audiometry in a lower noise environment.

[0119] Specifically, step S10312, namely adjusting the audiometry configuration of the hearing aid according to the signal-to-noise ratio, includes:

[0120] Step S103121: Obtain the preset signal-to-noise ratio threshold corresponding to the signal-to-noise ratio;

[0121] Step S103122: If the signal-to-noise ratio is higher than the preset signal-to-noise ratio threshold, then reduce the noise reduction depth in the audiometry configuration;

[0122] Step S103123: If the signal-to-noise ratio is lower than the preset signal-to-noise ratio threshold, then increase the noise reduction depth in the audiometry configuration.

[0123] In this embodiment of the invention, after determining that the sound field environment is suitable for hearing testing, the noise reduction depth of the hearing aid is adjusted based on the signal-to-noise ratio (SNR) and a preset SNR threshold. SNR is an important indicator of signal quality, reflecting the relative intensity of the signal to background noise. By adjusting the noise reduction depth, the output of the hearing aid can be optimized, ensuring that the user can obtain a clear auditory experience in various sound field environments.

[0124] The preset signal-to-noise ratio threshold is set based on the hearing aid's performance and the user's hearing needs to ensure optimal auditory performance in various environments. Through these steps, the hearing aid can automatically adjust its audiometric configuration according to the real-time sound field environment and noise conditions, thereby providing personalized hearing support for the user.

[0125] In this embodiment, by comparing the signal-to-noise ratio (SNR) with a preset SNR threshold, the hearing aid can automatically adjust its noise reduction function to adapt to different environmental noise levels. When the SNR is high, it indicates that the signal in the environment is relatively strong and the noise impact is small. Therefore, the noise reduction depth can be appropriately reduced to maintain the naturalness and clarity of the signal. Conversely, if the SNR is low, it indicates that noise has a greater impact on the signal. In this case, increasing the noise reduction depth helps to reduce noise interference and improve speech intelligibility. Through this dynamic adjustment, the hearing aid can provide users with a clearer and more comfortable auditory experience, while ensuring optimal hearing performance in various sound field environments.

[0126] In practical applications, for example, if the sound field environment is suitable for audiometry and the signal-to-noise ratio (SNR) is 10dB, which is lower than the preset SNR threshold of 15dB, the hearing aid will automatically increase the noise reduction depth to enhance the speech signal and reduce background noise interference. This allows users to have a clearer auditory experience even in noisy environments. This adaptive adjustment mechanism of the hearing aid ensures optimal auditory support in different noise environments, meeting the user's personalized needs.

[0127] In some alternative embodiments, after acquiring the environmental noise energy at key frequencies in the sound spectrum, the method further includes:

[0128] Based on the frequency (i.e., noise frequency) and intensity corresponding to the ambient noise energy, the masking threshold of the target signal of the ambient sound in each frequency band is calculated; the target signal is the audio signal used for audiometry, that is, a signal in a specific frequency band; specifically, the target signal refers to the sound signal that the hearing aid user wants to hear, usually referring to important signals such as speech or music; the intensity of ambient noise energy is usually expressed in the form of power value, and power spectral density or power value can more directly reflect the energy distribution of the signal; the frequency and intensity corresponding to the ambient noise energy are the noise frequency and the corresponding intensity.

[0129] The target signal of the ambient sound is compared with the corresponding masking threshold, and the audibility of the target signal of the ambient sound is corrected based on the comparison result.

[0130] In this embodiment of the invention, the step of correcting the audibility of the target signal of ambient sound can be performed only if it is determined that the current environment is suitable for listening tests. If it is determined that the current environment is not suitable for listening tests, then it will not be performed. For the process of determining whether the current environment is suitable for listening tests, please refer to the above.

[0131] In this embodiment, the masking threshold refers to the threshold for determining whether environmental noise energy affects the audibility of a target signal at a specific frequency. The masking effect is a phenomenon in human hearing perception, where a stronger sound can mask a weaker sound, making the latter difficult to detect. By calculating the masking threshold, the minimum intensity the target signal needs to achieve in each frequency band under the current environmental noise level can be determined to ensure its audibility. Specifically, the energy distribution and frequency characteristics of the environmental noise, as well as the frequency characteristics of the target signal, need to be considered when calculating the masking threshold. After calculating the masking threshold, the target signal can be adjusted accordingly to ensure its audibility in noisy environments. For example, if the environmental noise energy in a certain frequency band is high, the masking threshold of the target signal in that band will also increase accordingly, requiring an increase in the intensity of the signal in that band to overcome the noise's influence. In this way, hearing aids can provide users with a clearer and more comfortable auditory experience.

[0132] The target signal refers to the sound the user wants to hear, typically an important signal such as speech or music. In this embodiment, the target signal refers to the speech signal corresponding to the audiometry test. Calculating the masking threshold of the target signal is crucial for optimizing the auditory experience. By accurately calculating the masking threshold, the hearing aid can adjust the frequency response of the output signal, thereby highlighting the target signal in noisy environments and reducing background noise interference. For example, if the target signal is masked by ambient noise in a certain frequency band, the hearing aid can increase the gain in that frequency band, making the target signal more prominent, thus enabling the user to more easily identify and understand the sound content.

[0133] The masking threshold can be calculated in the following way:

[0134] First, the energy distribution of environmental noise is analyzed: This requires preprocessing the ambient sound to extract Mel-frequency cepstral coefficients (MFCCs), generating corresponding audio feature vectors, and then classifying and evaluating the sound field environment using a neural network. Next, based on the sound spectrum data of this sound field environment, the noise energy values ​​at each key frequency point are calculated, thus obtaining the environmental noise energy at the key frequencies in the sound spectrum.

[0135] Next, determine the frequency characteristics of the target signal: the target signal is typically the sound the user wants to hear, such as speech or music. Understanding the frequency characteristics of the target signal helps predict which frequency bands might be affected by noise.

[0136] Next, psychoacoustic models are applied: the masking threshold can be calculated using psychoacoustic models. Psychoacoustic models can simulate the auditory characteristics of the human ear, taking into account the influence of masking effects. For example, when a pure tone is masked by continuous noise with a certain bandwidth centered on it, if the power of the pure tone when it is just heard is equal to the noise power within that frequency band, then this bandwidth is the auditory critical bandwidth.

[0137] Finally, the masking threshold is calculated: based on the above information, combined with the characteristics of environmental noise energy and target signal, the masking threshold for each frequency band is calculated.

[0138] Correcting the audibility of the target signal for ambient sound is mainly achieved by adjusting the gain of the target signal in each frequency band:

[0139] First, the target signal is matched against the masking threshold: the target signal is compared with the calculated masking threshold. If the strength of the target signal in a certain frequency band is lower than the masking threshold for that band, it means that the target signal in that band may not be audible to the user and needs to be enhanced.

[0140] Then, adjust the target signal gain: based on the comparison results, appropriately increase the gain of the target signal in the affected frequency band so that its strength exceeds the masking threshold, thereby ensuring that the user can hear the target signal clearly.

[0141] Adjusting the gain of the target signal in each frequency band also falls under the category of adjusting the audiometry configuration of the hearing aid.

[0142] Other optional embodiments of the hearing aid audiometry method provided in this embodiment also include:

[0143] If multiple sounds are mixed in the sound field environment, the direction of the test sound source is determined based on a directionality estimation algorithm, and the signal reception intensity in the corresponding direction is enhanced. The test sound source is the source (sound-emitting device) that emits the audio used for the test.

[0144] Controlling and enhancing the signal reception strength in the corresponding direction also falls under the category of adjusting the hearing aid's audiometric configuration.

[0145] In this embodiment, if the sound field environment is reverberant and there is a need for sound from a specific direction, such as in a conference room, the hearing aid can utilize a directional estimation algorithm to identify and enhance sounds from that specific direction. This directional estimation algorithm analyzes sound signals from different directions within the sound field and enhances the sound source of interest to the user by adjusting the hearing aid's receiving mode. For example, if the user wants to focus on the voice of a speaker in front, the hearing aid can reduce interference from other directions, such as noise from the sides or rear, thereby improving the clarity and intelligibility of the target sound. Furthermore, the directional algorithm can help users better locate sound sources in noisy environments, which is crucial for understanding multi-person conversations or identifying specific sounds in complex sound fields.

[0146] The following methods can be used to determine whether multiple sounds are mixed in a sound field environment:

[0147] 1. Detection using a microphone array

[0148] Hearing aids are typically equipped with multiple microphones, which can be pointed in different directions to receive signals. By comparing the signal strength and phase difference received by different microphones, it is possible to make a preliminary determination of whether there are multiple sound sources.

[0149] If the signals received by different microphones have significant time delays or phase differences, it may indicate the presence of multiple sound sources. For example, the time delay between two microphones can be estimated using the Generalized Cross-Correlation (GCC) method.

[0150] 2. Spectrum Analysis

[0151] Perform spectral analysis on the received audio signal to check for multiple frequency components or uneven energy distribution within the frequency band. Multiple peaks in the spectrum may indicate the presence of multiple sound sources.

[0152] Using the Short-Time Fourier Transform (STFT) or other time-frequency analysis tools can help identify multiple frequency components in a signal.

[0153] 3. Energy detection

[0154] By monitoring changes in signal energy, the presence of multiple sound sources can be detected. If the signal energy fluctuates significantly over a period of time, it may indicate that multiple sound sources are emitting sound alternately.

[0155] Energy detection can be achieved by calculating the square of the signal and averaging it, a method that can quickly respond to changes in the sound source.

[0156] In addition, the direction of a sound source can be determined in the following ways:

[0157] 1. Methods based on Time Difference of Arrival (TDOA)

[0158] Time difference of arrival (TDOA) is one of the most commonly used techniques for sound source localization. By measuring the time difference between the arrival of sound waves at different microphones, the direction of the sound source can be calculated.

[0159] 2. Generalized Cross-Correlation (GCC) Method

[0160] The generalized cross-correlation (GCC) method is an improved time delay estimation technique that suppresses the effects of noise and reverberation by weighting the cross-power spectrum in the frequency domain.

[0161] GCC-PHAT (Phase Transform) is a variant of GCC that makes the GCC in the time domain sharper by normalizing the amplitude of the cross spectrum, which helps to estimate the time delay more accurately.

[0162] 3. MUSIC Algorithm

[0163] Multiple Signal Classification (MUSIC) is a high-resolution Direction of Arrival (DOA) estimation method, particularly suitable for working in high-noise environments. The MUSIC algorithm utilizes subspace decomposition techniques, leveraging the abundance of known signals to effectively separate the signal from background noise and accurately capture the directional information of each sound source.

[0164] The hearing aid testing method provided in this embodiment acquires the ambient sound of the user's current environment, determines the signal-to-noise ratio (SNR) of the ambient sound, adjusts the hearing aid's testing configuration according to the SNR, and controls the hearing aid to perform hearing tests according to the adjusted configuration. In other words, the hearing aid can automatically adjust its testing configuration based on the real-time sound field environment and noise conditions, enabling hearing tests for hearing aid users under different noise conditions. This improves the accuracy and reliability of the test results, adapts to the user's actual environment, provides personalized hearing support, and ultimately provides users with clearer and more accurate sound enhancement effects.

[0165] This embodiment also provides a hearing aid testing device for implementing the above embodiments and preferred embodiments; details already described will not be repeated. As used below, the term "module" can refer to a combination of software and / or hardware that performs a predetermined function. Although the device described in the following embodiments is preferably implemented in software, hardware implementation, or a combination of software and hardware, is also possible and contemplated.

[0166] This embodiment provides a hearing test device for a hearing aid, such as... Figure 2 As shown, it includes:

[0167] The ambient sound acquisition module 201 is used to acquire the ambient sound of the current environment.

[0168] Signal-to-noise ratio determination module 202 is used to determine the signal-to-noise ratio corresponding to the ambient sound;

[0169] The hearing test module 203 is used to adjust the hearing test configuration of the hearing aid according to the signal-to-noise ratio, and control the hearing aid to perform hearing tests according to the hearing test configuration.

[0170] In some optional embodiments, the hearing aid's audiometry device further includes:

[0171] The sound spectrum acquisition module is used to acquire the sound spectrum corresponding to the ambient sound;

[0172] An environmental noise energy acquisition module is used to acquire the environmental noise energy at key frequencies in the sound spectrum, wherein the key frequencies are predetermined hearing aid audiometry frequencies.

[0173] The audiometry module 203 is specifically used to adjust the audiometry configuration of the hearing aid based on the ambient noise energy and the signal-to-noise ratio.

[0174] In some optional embodiments, the environmental noise energy acquisition module includes:

[0175] The acquisition unit is used to acquire the average power value of the power spectrum at each time point and the power value corresponding to the key frequency, wherein the power spectrum at each time point is the power spectrum of each short frame of the ambient sound segmentation;

[0176] A noise discrimination threshold setting unit is used to set a noise discrimination threshold based on the average power value;

[0177] The first comparison unit is used to compare the power value corresponding to each of the key frequencies with the noise discrimination threshold.

[0178] A noise frequency identification unit is used to identify key frequencies whose power values ​​exceed the noise discrimination threshold as noise frequencies.

[0179] An environmental noise energy determination unit is used to determine the environmental noise energy based on the power value of the noise frequency.

[0180] In some optional embodiments, the audiometry module 203 includes:

[0181] The second comparison unit is used to obtain the magnitude relationship between the environmental noise energy and the preset noise energy threshold.

[0182] The noise reduction depth adjustment unit is used to adjust the noise reduction depth in the audiometry configuration according to the signal-to-noise ratio if it is determined that the current environment is suitable for audiometry based on the size relationship.

[0183] In some optional embodiments, the audiometry module 203 includes:

[0184] A signal-to-noise ratio threshold acquisition unit is used to acquire a preset signal-to-noise ratio threshold corresponding to the signal-to-noise ratio;

[0185] The noise reduction unit is used to reduce the noise reduction depth in the audiometry configuration if the signal-to-noise ratio is higher than the preset signal-to-noise ratio threshold.

[0186] An enhancement unit is used to enhance the noise reduction depth in the audiometry configuration if the signal-to-noise ratio is lower than the preset signal-to-noise ratio threshold.

[0187] In some optional embodiments, the hearing aid's audiometry device further includes:

[0188] The masking threshold calculation module is used to calculate the masking threshold of the target signal of the ambient sound in each frequency band based on the frequency and intensity corresponding to the ambient noise energy; the target signal is the audio signal used for listening tests.

[0189] The audibility correction module is used to compare the target signal of the ambient sound with the corresponding masking threshold, and correct the audibility of the target signal of the ambient sound based on the comparison result.

[0190] In some optional embodiments, the sound spectrum acquisition module includes:

[0191] The sound field environment acquisition unit is used to acquire the sound field environment corresponding to the ambient sound using a neural network model.

[0192] The sound spectrum acquisition unit is used to acquire the sound spectrum corresponding to the sound field environment as the sound spectrum corresponding to the ambient sound.

[0193] In some optional implementations, the sound field environment acquisition unit includes:

[0194] An extraction subunit is used to extract the Mel frequency cepstral coefficients of the ambient sound;

[0195] The feature generation subunit is used to generate the corresponding audio feature vector based on the Mel frequency cepstral coefficients;

[0196] The classification and prediction subunit is used to classify and evaluate the audio feature vector using the neural network model to obtain the corresponding sound field environment.

[0197] In some optional implementations, the extraction subunit is specifically used to segment the ambient sound into multiple short-time frames; apply a preset window function to each short-time frame; perform a fast Fourier transform on each short-time frame to calculate the power spectrum of each short-time frame; pass the power spectrum of each short-time frame through a set of Mel filter banks to generate multiple filter outputs; take the logarithm of the energy of any filter output, and perform a discrete cosine transform on the logarithmic energy value to generate cepstral coefficients; select the first N cepstral coefficients as the Mel frequency cepstral coefficients of the ambient sound.

[0198] In some optional embodiments, the hearing aid's audiometry device further includes:

[0199] The enhancement module is used to determine the direction of the test sound source based on a directionality estimation algorithm if there are multiple mixed sounds in the sound field environment, and to control the enhancement of the signal reception intensity in the corresponding direction.

[0200] Further functional descriptions of the above modules and units are the same as those in the corresponding embodiments described above, and will not be repeated here.

[0201] In this embodiment, the hearing aid's audiometry device is presented in the form of a functional unit. Here, a unit refers to an ASIC (Application Specific Integrated Circuit) circuit, a processor and memory that execute one or more software or fixed programs, and / or other devices that can provide the above-mentioned functions.

[0202] This invention also provides a hearing aid having the above-described features. Figure 2 The hearing aid's audiometry device is shown.

[0203] Please see Figure 3 , Figure 3 This is a schematic diagram of the structure of a hearing aid provided in an optional embodiment of the present invention, such as... Figure 3 As shown, the hearing aid includes one or more processors 10, a memory 20, and interfaces for connecting the various components, including high-speed interfaces and low-speed interfaces. The various components communicate with each other via different buses and can be mounted on a common motherboard or otherwise installed as needed. The processors can process instructions executed within the hearing aid. In some alternative embodiments, multiple processors and / or multiple buses can be used with multiple memories and multiple memory sets, if desired. Figure 3 Take a processor 10 as an example.

[0204] Processor 10 may be a central processing unit, a network processor, or a combination thereof. Processor 10 may further include a hardware chip. The hardware chip may be an application-specific integrated circuit (ASIC), a programmable logic device (PLD), or a combination thereof. The programmable logic device may be a complex programmable logic device (CAMP), a field-programmable gate array (FPGA), a general-purpose array logic (GDA), or any combination thereof.

[0205] The memory 20 stores instructions executable by at least one processor 10 to cause the at least one processor 10 to perform the method shown in the above embodiments.

[0206] The memory 20 may include a program storage area and a data storage area. The program storage area may store the operating system and applications required for at least one function; the data storage area may store data created based on the use of the hearing aid. Furthermore, the memory 20 may include high-speed random access memory and may also include non-transitory memory, such as at least one disk storage device, flash memory device, or other non-transitory solid-state storage device. In some alternative embodiments, the memory 20 may optionally include memory remotely located relative to the processor 10, which can be connected to the hearing aid via a network. Examples of such networks include, but are not limited to, the Internet, corporate intranets, local area networks, mobile communication networks, and combinations thereof.

[0207] The memory 20 may include volatile memory, such as random access memory; the memory may also include non-volatile memory, such as flash memory, hard disk or solid-state drive; the memory 20 may also include a combination of the above types of memory.

[0208] The hearing aid also includes an input device 30 and an output device 40. The processor 10, memory 20, input device 30, and output device 40 can be connected via a bus or other means. Figure 3 Taking a bus connection as an example, the input device may include, for example, a microphone for collecting ambient sound and converting it into an electrical signal. The hearing aid may also include an amplifier for receiving the electrical signals converted by the microphone and processing and amplifying these signals. Depending on the user's hearing loss characteristics, the amplifier can specifically adjust the sound intensity at different frequencies. The output device 40 may include a speaker for converting the amplified electrical signal back into an audio signal and transmitting it to the user's ear.

[0209] The hearing aid may also include a communication interface for communicating with other devices or communication networks.

[0210] This invention also provides a computer-readable storage medium. The methods described above according to embodiments of the invention can be implemented in hardware or firmware, or implemented as computer code that can be recorded on a storage medium, or implemented as computer code downloaded via a network and originally stored on a remote storage medium or a non-transitory machine-readable storage medium and then stored on a local storage medium. Thus, the methods described herein can be processed by software stored on a storage medium using a general-purpose computer, a dedicated processor, or programmable or dedicated hardware. The storage medium can be a magnetic disk, optical disk, read-only memory, random access memory, flash memory, hard disk, or solid-state drive, etc.; further, the storage medium can also include combinations of the above types of memory. It is understood that computers, processors, microprocessor controllers, or programmable hardware include storage components capable of storing or receiving software or computer code, which, when accessed and executed by the computer, processor, or hardware, implements the methods shown in the above embodiments.

[0211] A portion of this invention can be applied as a computer program product, such as computer program instructions, which, when executed by a computer, can invoke or provide the methods and / or technical solutions according to the invention through the operation of the computer. Those skilled in the art will understand that the forms in which computer program instructions exist in a computer-readable medium include, but are not limited to, source files, executable files, installation package files, etc. Correspondingly, the ways in which computer program instructions are executed by a computer include, but are not limited to: the computer directly executing the instructions, or the computer compiling the instructions and then executing the corresponding compiled program, or the computer reading and executing the instructions, or the computer reading and installing the instructions and then executing the corresponding installed program. Here, the computer-readable medium can be any available computer-readable storage medium or communication medium accessible to a computer.

[0212] Although embodiments of the invention have been described in conjunction with the accompanying drawings, those skilled in the art can make various modifications and variations without departing from the spirit and scope of the invention, and all such modifications and variations fall within the scope defined by the appended claims.

Claims

1. A hearing test method for a hearing aid, characterized in that, The method includes: Get the ambient sound of the current environment; Determine the signal-to-noise ratio corresponding to the ambient sound; Obtain the sound spectrum corresponding to the ambient sound; The method involves obtaining the environmental noise energy at key frequencies in the sound spectrum, where the key frequencies are predetermined hearing aid audiometry frequencies. This includes: obtaining the average power value of the power spectrum at each time point and the power value corresponding to the key frequency, where the power spectrum at each time point is the power spectrum of each short frame of the environmental sound segmentation; setting a noise discrimination threshold based on the average power value; comparing the power value corresponding to each key frequency with the noise discrimination threshold; identifying key frequencies with power values ​​exceeding the noise discrimination threshold as noise frequencies; and determining the environmental noise energy based on the power value of the noise frequencies. Based on the ambient noise energy and the signal-to-noise ratio, the hearing aid's audiometry configuration is adjusted, and the hearing aid is controlled to perform audiometry according to the audiometry configuration.

2. The method according to claim 1, characterized in that, The step of adjusting the audiometry configuration of the hearing aid based on the ambient noise energy and the signal-to-noise ratio includes: Obtain the relationship between the ambient noise energy and the preset noise energy threshold; If, based on the aforementioned size relationship, it is determined that the current environment is suitable for audiometry, then the noise reduction depth in the audiometry configuration is adjusted according to the signal-to-noise ratio.

3. The method according to claim 1 or 2, characterized in that, The step of adjusting the audiometry configuration of the hearing aid based on the signal-to-noise ratio includes: Obtain the preset signal-to-noise ratio threshold corresponding to the signal-to-noise ratio; If the signal-to-noise ratio is higher than the preset signal-to-noise ratio threshold, then the noise reduction depth in the audiometry configuration is reduced; If the signal-to-noise ratio is lower than the preset signal-to-noise ratio threshold, then the noise reduction depth in the audiometry configuration is increased.

4. The method according to claim 2, characterized in that, After obtaining the environmental noise energy at key frequencies in the sound spectrum, the method further includes: Based on the frequency and intensity corresponding to the ambient noise energy, the masking threshold of the target signal of the ambient sound in each frequency band is calculated; the target signal is the audio signal used for listening measurement. The target signal of the ambient sound is compared with the corresponding masking threshold, and the audibility of the target signal of the ambient sound is corrected based on the comparison result.

5. The method according to claim 1, characterized in that, The step of obtaining the sound spectrum corresponding to the ambient sound includes: The sound field environment corresponding to the ambient sound is obtained using a neural network model; The sound spectrum corresponding to the sound field environment is obtained as the sound spectrum corresponding to the ambient sound.

6. The method according to claim 5, characterized in that, The step of using a neural network model to obtain the sound field environment corresponding to the ambient sound includes: Extract the Mel-frequency cepstral coefficients of the ambient sound; Based on the Mel frequency cepstral coefficients, a corresponding audio feature vector is generated; The neural network model is used to classify and evaluate the audio feature vectors to obtain the corresponding sound field environment.

7. The method according to claim 6, characterized in that, The extraction of the Mel frequency cepstral coefficients of the ambient sound includes: The ambient sound is segmented into multiple short frames; A preset window function is applied to each of the aforementioned short frames; Perform a fast Fourier transform on each short-time frame for which the preset window function is applied, and calculate the power spectrum of each short-time frame; The power spectrum of each short frame is passed through a set of Mel filter banks to generate multiple filter outputs; Take the logarithm of the energy output of any filter, and perform a discrete cosine transform on the logarithmic energy value to generate cepstral coefficients; The first N cepstral coefficients are selected as the Mel frequency cepstral coefficients of the ambient sound.

8. The method according to claim 1, characterized in that, Also includes: If there are multiple mixed sounds in the sound field environment, the direction of the test sound source is determined based on the directionality estimation algorithm, and the signal reception intensity in the corresponding direction is enhanced.

9. A hearing test device for a hearing aid, characterized in that, The device includes: The ambient sound acquisition module is used to acquire the ambient sound of the current environment; The signal-to-noise ratio (SNR) determination module is used to determine the SNR corresponding to the ambient sound. The sound spectrum acquisition module is used to acquire the sound spectrum corresponding to the ambient sound; An environmental noise energy acquisition module is used to acquire the environmental noise energy at key frequencies in the sound spectrum, wherein the key frequencies are predetermined hearing aid audiometry frequencies. The environmental noise energy acquisition module includes: an acquisition unit, configured to acquire the average power value of the power spectrum at each time point and the power value corresponding to the key frequency, wherein the power spectrum at each time point is the power spectrum of each short frame of the environmental sound segmentation; a noise discrimination threshold setting unit, configured to set a noise discrimination threshold based on the average power value; a first comparison unit, configured to compare the power value corresponding to each key frequency with the noise discrimination threshold; a noise frequency identification unit, configured to identify key frequencies whose power values ​​exceed the noise discrimination threshold as noise frequencies; and an environmental noise energy determination unit, configured to determine the environmental noise energy based on the power value of the noise frequency. The audiometry module is used to adjust the audiometry configuration of the hearing aid according to the ambient noise energy and the signal-to-noise ratio, and to control the hearing aid to perform audiometry according to the audiometry configuration.

10. A hearing aid, characterized in that, include: A memory and a processor are communicatively connected, the memory stores computer instructions, and the processor executes the computer instructions to perform the audiometry method of the hearing aid as described in any one of claims 1 to 8.

11. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores computer instructions for causing the computer to perform the hearing test method of any one of claims 1 to 8.

Images