A portable precise hearing detection system based on adaptive sound field compensation
By using adaptive sound field compensation technology and employing microphone arrays and deep learning algorithms to construct a sound field environment parameter model, the problem of external interference in portable hearing testing systems is solved, thereby improving detection accuracy and efficiency.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- TONGJI HOSPITAL ATTACHED TO TONGJI MEDICAL COLLEGE HUAZHONG SCI TECH
- Filing Date
- 2025-10-21
- Publication Date
- 2026-06-16
Smart Images
Figure CN120998238B_ABST
Abstract
Description
Technical Field
[0001] This invention relates to the field of hearing testing technology, and in particular to a portable and precise hearing testing system based on adaptive sound field compensation. Background Technology
[0002] With the improvement of people's living standards and the enhancement of health awareness, there is an increasing focus on hearing health, especially for the elderly and workers who are exposed to noisy environments for extended periods, who need regular hearing monitoring. Furthermore, the rapid development of mobile healthcare and smart technology has provided technical support for the development of portable hearing monitoring devices.
[0003] A search revealed Chinese patent CN110960224A, which discloses a hearing threshold and / or hearing status detection system and method. The detection system includes: a data acquisition and transmission system for transmitting stimulus signals and acquiring ear canal signals; and a hearing threshold analysis and prediction system, including a hearing threshold detection module, a routine testing module, and / or a hearing status screening module. The hearing threshold detection module determines the hearing threshold corresponding to different stimulus frequencies using a pre-trained network model. The routine testing module adaptively selects a test intensity range through the data acquisition and transmission system and predicts the hearing threshold corresponding to the stimulus frequency point using the pre-trained network model. The screening module performs hearing status screening through the data acquisition and transmission system using the pre-trained network model. This invention not only provides accurate detection results but is also applicable to various application scenarios.
[0004] Compared with existing technologies, the invention patent with Chinese patent number CN110960224A can generate different stimulation frequencies and intensities according to the different test contents required by the test subject, thereby obtaining the hearing status of different test subjects and improving the accuracy of hearing status detection to a certain extent.
[0005] However, in actual use, the above system only generates different stimulus parameters for different test subjects based on the test content, without considering the test state during the test. This may lead to external interference affecting the test results and thus affecting the accuracy of the test. Summary of the Invention
[0006] The purpose of this invention is to address the shortcomings of insufficient accuracy in existing technologies by proposing a portable and precise hearing detection system based on adaptive sound field compensation.
[0007] To achieve the above objectives, the present invention adopts the following technical solution:
[0008] A portable, precise hearing detection system based on adaptive sound field compensation includes:
[0009] The environmental monitoring module is used to acquire the corresponding environmental noise signal from the corresponding environmental terminal within the portable hearing monitoring device, and to preprocess the environmental noise signal through the environmental terminal.
[0010] The hearing monitoring module is used to acquire corresponding audio test signals and user feedback information from the corresponding test detection end within the portable hearing monitoring device;
[0011] The data transmission module is used to set the environmental transmission channel protocol and the test transmission channel protocol corresponding to the data processing end in the portable hearing monitoring device, and to transmit the corresponding data information according to the corresponding transmission channel protocol.
[0012] The environmental compensation module is used to set an environmental sequence frame based on the acquisition time according to the environmental noise signal obtained within the environmental transmission channel protocol, obtain the initial sound field environmental parameters based on the corresponding environmental noise signal within the environmental sequence frame, perform environmental adaptive compensation on the audio test signal based on the initial sound field environmental parameters, and obtain the initial audio test signal.
[0013] The test compensation module is used to compare and analyze the obtained audio test signal with the initial audio test signal to obtain deviation audio test information, and to adaptively compensate the initial audio test signal based on the deviation audio test information to obtain a comprehensive audio test signal.
[0014] The detection and analysis module is used to obtain user feedback information corresponding to the comprehensive audio test signal, extract features from the user feedback information, construct a hearing test comparison chart, analyze and process the hearing test comparison chart, and obtain the corresponding hearing test data.
[0015] The above technical solution further includes: the environmental monitoring module includes:
[0016] The environmental monitoring and processing unit is used to acquire corresponding environmental noise signals from the corresponding microphone array at the environmental end, mark the environmental noise signals according to the corresponding microphone number and acquisition time, and decompose the marked environmental noise signals based on the independent component analysis method to obtain the corresponding independent source signals.
[0017] The environmental monitoring classification unit is used to extract features from historical source signals, obtain corresponding time-domain and frequency characteristics, and set corresponding classification intervals based on the feature extraction results; it also performs matching analysis between the source signals corresponding to the environmental noise signals and the corresponding classification intervals.
[0018] Furthermore, the hearing monitoring module includes:
[0019] The test monitoring and processing unit is used to acquire audio test signals from the ear of the tester, mark the acquired audio test signals according to the acquisition time, extract features from the audio test signals to obtain audio test feature data, and classify the audio test feature data.
[0020] The feedback monitoring unit is used by the detection end to acquire user feedback information from the tester during the test, and to mark the user feedback information according to the collection time.
[0021] Furthermore, the data transmission module includes:
[0022] The transmission management unit is used to obtain the classification intervals corresponding to the environment end and the detection end, set the corresponding environment classification interval protocol and test classification interval protocol according to the corresponding feature extraction results in each classification interval, sort the environment classification interval protocol and test classification interval protocol respectively, and construct the transmission channel protocol architecture on the sorting results. Based on the architecture results, the corresponding environment transmission channel protocol and test transmission channel protocol are generated respectively.
[0023] The distribution of microphone arrays in the environment and detection ends is obtained respectively. Based on the distribution of microphone arrays, corresponding environmental transmission sub-channels and test transmission sub-channels are set. The environmental transmission channel protocol and test transmission channel protocol are mapped to each environmental transmission sub-channel and test transmission sub-channel respectively.
[0024] The data transmission unit is used by the environmental end and the detection end to transmit the corresponding data information according to the environmental transmission sub-channel and the test transmission sub-channel connected to the data processing end, respectively, to obtain the matching analysis results of the corresponding data information with the classification interval protocol in sequence during the transmission process, and to generate classification matching codes based on the matching results.
[0025] Furthermore, the environmental compensation module includes:
[0026] The data preprocessing unit is used to obtain the historical environmental noise signals and historical sound field environmental parameters corresponding to the corresponding classification intervals, set the corresponding classification noise sound field datasets, and analyze and train the obtained classification noise sound field datasets based on deep learning algorithms to construct classification sound field environmental parameter models.
[0027] The model pre-matching unit is used to acquire the environmental noise signal transmitted from the environmental end and the corresponding classification matching code, acquire the corresponding classification sound field environmental parameter model according to the corresponding classification matching code, and analyze and process the acquired environmental noise signal through the corresponding classification sound field environmental parameter model to obtain the corresponding channel sound field environmental parameters.
[0028] The fusion processing unit is used to set up corresponding three-dimensional channel space models according to the environmental transmission sub-channels; to map the environmental noise signals and channel sound field environmental parameters obtained by the corresponding environmental transmission sub-channels into the corresponding three-dimensional channel space models for visualization processing, to obtain the corresponding three-dimensional environmental space images, and to set up sequence frames according to the acquisition time, to fuse the environmental noise signals and channel sound field environmental parameters corresponding to the corresponding pixels in each sequence frame corresponding to the same acquisition time, to obtain the initial sound field environmental parameters corresponding to the acquisition time.
[0029] An environmental compensation unit is used to acquire the initial audio test signal.
[0030] Furthermore, the process by which the environmental compensation unit acquires the initial audio test signal includes:
[0031] The initial sound field environment parameters and corresponding audio test signals corresponding to the acquisition time are obtained. An adaptive algorithm is selected based on the initial sound field environment parameters. The adaptive algorithm includes LMS, FDAF and beamforming algorithm. The environmental interference is iteratively compensated for based on the selected adaptive algorithm. The audio test signals are then adapted based on the compensated parameters to generate the initial audio test signals.
[0032] Furthermore, the test compensation module includes:
[0033] The audio comparison unit is used to obtain the classification interval corresponding to the initial audio test signal, as well as the classification matching code and corresponding test monitoring signal obtained by the test transmission sub-channel. The corresponding classification matching code is compared and analyzed with the corresponding classification interval in turn to determine whether they belong to the same classification interval. The initial audio test signal and test monitoring information that do not belong to the same classification interval are compared and analyzed first to obtain the corresponding deviation audio test signal.
[0034] The test fusion unit is used to set up corresponding three-dimensional test space models according to the test transmission sub-channels, map the test monitoring signals and deviation audio test signals obtained by the corresponding test transmission sub-channels into the three-dimensional test space models, obtain the corresponding three-dimensional test space images, and perform fusion processing on the deviation audio test signals at corresponding pixels in the three-dimensional test space images corresponding to the same acquisition time to obtain the deviation fusion signal corresponding to the acquisition time.
[0035] The test compensation unit is used to select the appropriate adaptive algorithm based on the obtained deviation fusion signal. The appropriate adaptive algorithm performs adaptive compensation on the initial audio test signal based on the deviation fusion signal to obtain the comprehensive audio test signal.
[0036] Furthermore, the detection and analysis module includes:
[0037] The image comparison unit is used to acquire the corresponding comprehensive audio test signal and user feedback information, and to compare and mark the comprehensive audio test signal with the user feedback information. The user feedback information includes subjective feedback information and otoacoustic reflex signal. Feature extraction is performed on the user feedback information to obtain feedback feature signals. The obtained feedback feature signals are used to establish hearing test comparison charts corresponding to the comprehensive audio test signal based on the corresponding time information.
[0038] The detection and analysis unit is used to analyze and process the hearing test comparison charts corresponding to each feedback feature signal, obtain the corresponding hearing test feature data, and perform comprehensive analysis on the obtained hearing test feature data to determine whether there are any contradictory hearing test feature data. If there are no contradictory hearing test feature data, the corresponding hearing test data is generated based on the comprehensive analysis results.
[0039] The present invention has the following beneficial effects:
[0040] 0. In this invention, by setting corresponding transmission sub-channels according to the microphone arrays corresponding to the environmental end and the test end, and by classifying and processing the obtained historical data information, setting corresponding classification interval protocols, the data information obtained by the environmental end and the test end is matched with the corresponding classification intervals during data transmission, thereby obtaining the corresponding classification matching code. Based on the obtained classification matching code, the corresponding classification sound field environment parameter model is prioritized for scheduling or comparative analysis is performed, thereby improving the efficiency of data analysis in the hearing test process and reducing the impact of delay on the test process.
[0041] 1. In this invention, by setting up three-dimensional spatial models for mapping processing of the data information obtained by the corresponding transmission sub-channels according to the distribution of the microphone arrays corresponding to the environmental end and the test end, the three-dimensional spatial images obtained by each mapping result are fused based on the corresponding pixels, the fusion processing results are integrated, and the environmental sound field in the space where the portable hearing monitoring device is located is adaptively compensated according to the integration result, thereby improving the accuracy of the hearing detection process.
[0042] 2. In this invention, by setting up an environmental compensation module and a test compensation module, adaptive sound field compensation is performed on the test environment of the corresponding portable hearing testing device. By setting up a dual-layer adaptive sound field compensation, the accuracy of adaptive sound field compensation during hearing testing is improved to a certain extent. Attached Figure Description
[0043] Figure 1 This is a schematic diagram of the structure of a portable precision hearing detection system based on adaptive sound field compensation proposed in this invention. Detailed Implementation
[0044] 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, and 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.
[0045] Example 1
[0046] like Figure 1 As shown, the present invention proposes a portable and precise hearing detection system based on adaptive sound field compensation, comprising:
[0047] The environmental monitoring module is used to acquire the corresponding environmental noise signal from the corresponding environmental terminal within the portable hearing monitoring device, and to preprocess the environmental noise signal through the environmental terminal.
[0048] The hearing monitoring module is used to acquire corresponding audio test signals and user feedback information from the corresponding test detection end within the portable hearing monitoring device;
[0049] The data transmission module is used to set the environmental transmission channel protocol and the test transmission channel protocol corresponding to the data processing end in the portable hearing monitoring device, and to transmit the corresponding data information according to the corresponding transmission channel protocol.
[0050] The environmental compensation module is used to set an environmental sequence frame based on the acquisition time according to the environmental noise signal obtained within the environmental transmission channel protocol, obtain the initial sound field environmental parameters based on the corresponding environmental noise signal within the environmental sequence frame, perform environmental adaptive compensation on the audio test signal based on the initial sound field environmental parameters, and obtain the initial audio test signal.
[0051] The test compensation module is used to compare and analyze the obtained audio test signal with the initial audio test signal to obtain deviation audio test information, and to adaptively compensate the initial audio test signal based on the deviation audio test information to obtain a comprehensive audio test signal.
[0052] The detection and analysis module is used to obtain user feedback information corresponding to the comprehensive audio test signal, extract features from the user feedback information, construct a hearing test comparison chart, analyze and process the hearing test comparison chart, and obtain the corresponding hearing test data.
[0053] In summary, the portable hearing testing device involved in this invention includes an environmental end, a testing end, and a data processing end. The environmental end analyzes and processes environmental data of the portable hearing testing device. The testing end monitors, collects, and processes feedback data from the user corresponding to the portable hearing testing device, and preprocesses the monitoring and collection results. The data processing end adaptively adjusts the data based on the preprocessing results from the environmental end and the testing end, facilitating more accurate acquisition of the user's hearing test results. This system completes the hearing test process for the user based on the environmental end, the testing end, and the data processing end corresponding to the portable hearing testing device, as well as the processing results between each end, thereby improving the timeliness of adaptive sound field compensation and the accuracy of the hearing test process to a certain extent.
[0054] In practical implementation, the environmental monitoring module includes:
[0055] An environmental monitoring and processing unit is used to acquire the corresponding microphone array of the corresponding environment in the portable hearing testing device. The microphone array includes corresponding microphone units, arrangement topology, and signal processing algorithms. The unit acquires the corresponding environmental noise signal based on the corresponding microphone array and marks the environmental noise signal according to the corresponding microphone unit and acquisition time.
[0056] The obtained environmental noise signal is filtered and windowed.
[0057] The corresponding environmental noise signals are decomposed and processed using the independent component analysis method to obtain the corresponding independent source signals;
[0058] Obtain the historical source signals corresponding to the independent component analysis method, extract features from the historical source signals respectively, and obtain the time-domain characteristics and frequency characteristics of the corresponding historical source signals;
[0059] The environmental monitoring classification unit is used to acquire the corresponding historical source signals, sort them according to their time-domain and frequency-domain characteristics, set corresponding classification intervals based on the sorting results, and statistically process the historical source signals within the corresponding classification intervals to obtain the historical source signals of the corresponding classification intervals.
[0060] The source signals corresponding to the corresponding environmental noise signals are compared and analyzed with the historical source signals corresponding to each classification interval to obtain the classification interval to which each source signal to which the corresponding environmental noise signal belongs.
[0061] It should be further explained that the sensor corresponding to the environmental end is a microphone array. The acquired environmental noise data is analyzed and processed by the microphone array according to the time difference, intensity difference and phase difference of the corresponding sound signal to obtain the source direction and signal characteristics of the corresponding environmental noise data. Based on beamforming technology, the source direction and signal characteristics of the corresponding environmental noise data are analyzed and processed to obtain the environmental noise signal corresponding to the corresponding environmental noise data. The corresponding environmental noise signal is then marked according to the location information of the corresponding microphone unit.
[0062] In practical implementation, the hearing monitoring module includes:
[0063] The test monitoring and processing unit is used to acquire the corresponding microphone array set at the ear of the tester, acquire the corresponding audio test data of the tester, analyze and process the audio test data, acquire the corresponding audio test signal, mark the acquired audio test signal according to the acquisition time, extract features from the audio test signal, acquire audio test feature data, and classify the audio test feature data. The classification process is consistent with the processing process in the environmental monitoring module.
[0064] The feedback monitoring unit is used to acquire user feedback information from the tester during the test and to mark the user feedback information according to the acquisition time. The user feedback information includes subjective feedback information and otoacoustic reflex signals. The subjective feedback information includes corresponding perceptual feedback, quality feedback, and difference feedback. The user fills in the corresponding feedback information according to the corresponding audio test signal. Perceptual feedback is used to directly determine whether the user has received the test signal, including feedback options such as "heard" or "did not hear". Quality feedback is used to evaluate the user's subjective feeling about the signal, including loudness evaluation: "too soft", "just right (comfortable)", "too loud" or "harsh", clarity evaluation: "completely clear content" or "blurry but words can be distinguished", comfort evaluation: "no discomfort" or "ear pain", and difference feedback includes time difference: "there is a pause between two sounds" or "it is a continuous sound", and masking difference: "the sound is clearer than before" or "completely covered by noise", etc. The user feedback options are recorded to generate user feedback information.
[0065] In specific implementation, the data transmission module includes:
[0066] The transmission management unit is used to obtain the classification intervals corresponding to the environment end and the detection end, set the corresponding environment classification interval protocol and test classification interval protocol according to the corresponding feature extraction results in each classification interval, sort the environment classification interval protocol and test classification interval protocol respectively, and construct the transmission channel protocol architecture on the sorting results. Based on the architecture results, the corresponding environment transmission channel protocol and test transmission channel protocol are generated respectively.
[0067] In practical implementation, taking the setup process of the environmental transmission channel protocol as an example, the process includes:
[0068] The process of setting up the environmental transmission channel protocol is as follows: acquire the environmental noise signal obtained in the corresponding environmental terminal, and set up the corresponding environmental transmission sub-channel according to the microphone unit in the microphone array corresponding to the corresponding sensor in the corresponding portable hearing device.
[0069] Obtain the classification interval of the historical source signal corresponding to the corresponding environmental noise signal, and set the corresponding classification interval protocol in sequence according to the corresponding classification interval. The classification interval protocol is the source signal feature information corresponding to the corresponding classification interval.
[0070] Statistical analysis is performed on the classification intervals to which the corresponding historical source signals belong, and the proportion data of each classification interval is obtained. The corresponding classification interval protocols are sorted according to the corresponding proportion data.
[0071] The sorting results of the classification interval protocol are integrated, the transmission channel protocol architecture is constructed based on the integration results, and the environmental transmission channel protocol is generated based on the transmission channel protocol architecture results.
[0072] The data transmission unit is used for the environmental end and the detection end to transmit the corresponding data information according to the environmental transmission sub-channel and the test transmission sub-channel connected to the data processing end, respectively, to obtain the matching analysis results of the corresponding data information with the classification interval protocol in sequence during the transmission process, and to generate classification matching codes based on the matching results;
[0073] Taking the environmental transmission subchannel as an example, the process of transmitting the corresponding environmental noise signal in the corresponding environmental transmission subchannel includes:
[0074] The corresponding ambient noise signal is input to the corresponding ambient transmission sub-channel according to the microphone array corresponding to the corresponding ambient noise signal. The ambient noise signal is analyzed and processed according to the ambient transmission channel protocol corresponding to each ambient transmission sub-channel. The source signal corresponding to the ambient noise signal is traversed and matched according to the corresponding classification interval protocol to obtain the matching result between the source signal and the corresponding classification interval protocol. The successfully matched classification interval protocol is marked and the corresponding classification matching code is generated.
[0075] The obtained classification matching code is acquired, marked according to its corresponding environmental transmission sub-channel, and transmitted to the data processing terminal in real time.
[0076] In practical implementation, the environmental compensation module includes:
[0077] The data preprocessing unit is used to acquire historical sound field environmental parameter information and corresponding historical environmental noise signals. Based on the classification matching code corresponding to the historical environmental noise signals, it sets up the corresponding classification noise sound field dataset, analyzes and processes the acquired classification noise sound field dataset, analyzes and trains the corresponding classification noise sound field datasets based on deep learning algorithms, constructs the classification sound field environmental parameter model based on the analysis and training results, and verifies and analyzes the classification sound field environmental parameter model after the initial training is completed until the corresponding loss function tends to stabilize. The corresponding classification sound field environmental parameter model is output and correlated with the corresponding classification matching code.
[0078] The model pre-matching unit is used to acquire the environmental noise signal transmitted from the environmental end and the corresponding classification matching code, acquire the corresponding classification sound field environmental parameter model according to the corresponding classification matching code, and analyze and process the acquired environmental noise signal through the corresponding classification sound field environmental parameter model to obtain the corresponding channel sound field environmental parameters.
[0079] The fusion processing unit is used to set up corresponding three-dimensional channel space models for each of the environmental transmission sub-channels.
[0080] The environmental noise signal and channel sound field environmental parameters obtained from the corresponding environmental transmission sub-channel are mapped into the corresponding three-dimensional channel space model for visualization processing to obtain the corresponding three-dimensional environmental space image.
[0081] Sequence frames are set according to the acquisition time of the corresponding environmental noise signal, and the sequence frame is a three-dimensional environmental space image corresponding to the environmental noise signal corresponding to the acquisition time.
[0082] Based on the corresponding sequence frames, the corresponding three-dimensional channel spatial models are mapped to the corresponding pixel positions in the three-dimensional environment spatial image. Based on the mapping results, the environmental noise signals and channel sound field environmental parameters in the corresponding three-dimensional channel spatial models are obtained.
[0083] The process of fusing the environmental noise signals and channel sound field parameters corresponding to each pixel includes:
[0084] Based on the position information of the corresponding microphone unit in the microphone array, obtain the spatial direction parameter information between the corresponding environmental transmission sub-channels, and obtain the historical environmental noise signal and historical sound field environmental parameters of the corresponding classification interval in the microphone array corresponding to the corresponding environmental transmission sub-channel.
[0085] The specific implementation process of obtaining the time difference of the corresponding sound source signal based on the corresponding historical environmental noise signal includes:
[0086] The corresponding time difference function is labeled as ,in, The time delay information used to reflect the signal at the corresponding pixel is obtained through the following formula:
[0087] ,in, Let x be the cross-power spectral density of microphone unit x and microphone unit y. Let t be the weighting function, and t be the corresponding time delay.
[0088] Obtain the azimuth angle between the corresponding microphone units , Where d is the spacing between the corresponding microphone units, c is the speed of sound propagation, and t is the corresponding time delay information;
[0089] Spatial correlation information is obtained based on the corresponding historical environmental noise signals. It can be obtained through the following formula:
[0090] ;
[0091] Determine whether spatial correlation information exists between corresponding microphone units within the microphone array and corresponding pixel within the corresponding classification interval. Verify and analyze the spatial correlation information based on the corresponding historical environmental noise signal. If it matches, output the corresponding spatial correlation information. ;
[0092] The corresponding pixels within the corresponding sequence frame are fused with the corresponding visualization processing results based on the corresponding spatial correlation information. The corresponding weight factors are set according to the spatial correlation information between the corresponding microphone units under the corresponding classification interval conditions. The environmental noise signal and channel sound field environmental parameters within the corresponding pixels are fused based on the corresponding weight factors to obtain the fused environmental noise signal and fused sound field environmental parameter information of the corresponding pixels.
[0093]
[0094]
[0095]
[0096]
[0097] A sound field environment parameter model is constructed. Based on the corresponding sound field environment parameter model, the corresponding fused environmental noise signal and pixel fused sound field environment parameter information are verified. The pixel fused sound field environment parameter information that meets the verification process is integrated in the three-dimensional environmental space image. The initial sound field environment parameters of the corresponding sequence frame are obtained according to the integration result.
[0098] The initial sound field environment parameters and corresponding audio test signals corresponding to the acquisition time are obtained. An adaptive algorithm is selected based on the initial sound field environment parameters. The adaptive algorithm includes LMS, FDAF and beamforming algorithm. The environmental interference is iteratively compensated for based on the selected adaptive algorithm. The audio test signals are then adapted based on the compensated parameters to generate the initial audio test signals.
[0099] In specific implementation, the test compensation module includes:
[0100] The audio comparison unit is used to acquire the classification interval corresponding to the initial audio test signal, as well as the classification matching code and corresponding test monitoring signal obtained from the test transmission sub-channel. It compares the corresponding classification matching code with the corresponding classification interval in sequence to determine whether they belong to the same classification interval. If they do not belong to the same classification interval, it sets a priority and prioritizes comparing and analyzing the corresponding initial audio test signal and test monitoring information to acquire the corresponding deviation audio test signal. If they belong to the same classification interval, it compares and analyzes the corresponding initial audio test signal and test monitoring information according to the acquisition order to acquire the corresponding deviation audio test signal.
[0101] The test fusion unit is used to set up corresponding three-dimensional test space models according to the test transmission sub-channels, map the test monitoring signals and deviation audio test signals obtained by the corresponding test transmission sub-channels into the three-dimensional test space models, obtain the corresponding three-dimensional test space images, and perform fusion processing on the deviation audio test signals at corresponding pixels in the three-dimensional test space images corresponding to the same acquisition time to obtain the deviation fusion signal corresponding to the acquisition time.
[0102] The test compensation unit is used to select the appropriate adaptive algorithm based on the obtained deviation fusion signal, and the appropriate adaptive algorithm performs adaptive compensation on the initial audio test signal based on the deviation fusion signal to obtain the comprehensive audio test signal.
[0103] It should be further explained that the specific implementation process of the test fusion unit and the test compensation unit is roughly the same as that of the fusion processing unit and the environmental compensation unit in the environmental compensation module. The difference lies in the fact that by analyzing and processing the historical audio test signals and historical sound field parameters corresponding to different classification intervals, the fusion correlation between corresponding pixels is determined, and corresponding weight factors are set to obtain the corresponding deviation fusion signal. Adaptive compensation is then performed based on the deviation fusion signal. By setting the test compensation module, the significant deviation between the tester and the initial audio test signal preset by the data processing terminal during the test process can be avoided, which affects the corresponding test process and thus improves the accuracy of the hearing test.
[0104] In specific implementation, the detection and analysis module includes:
[0105] An image comparison unit is used to acquire corresponding comprehensive audio test signals and user feedback information, and to compare and label the comprehensive audio test signals with the user feedback information. The user feedback information includes subjective feedback information and otoacoustic reflex signals. Feature extraction is performed on the user feedback information to obtain feedback feature signals. Based on the corresponding time information, hearing test comparison images corresponding to the comprehensive audio test signals are established using the obtained feedback feature signals.
[0106] The hearing test comparison chart can include the test signal parameters corresponding to the comprehensive audio test signal corresponding to the user feedback information, such as frequency and sound pressure level;
[0107] The detection and analysis unit is used to perform hearing detection analysis, such as hearing threshold analysis and frequency response analysis, on the hearing test comparison charts corresponding to each feedback feature signal to obtain the corresponding hearing test feature data. The obtained hearing test feature data is then combined in pairs, and the results of the pairwise combinations are logically analyzed to determine whether there are any contradictory hearing test feature data. If there are no contradictory hearing test feature data, the corresponding hearing test data is generated based on the comprehensive analysis results. If there are contradictory hearing test feature data, hearing test abnormality information is generated, and the corresponding user is retested.
[0108] Although embodiments of the invention have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions and variations can be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the appended claims and their equivalents.
Claims
1. A portable, precise hearing detection system based on adaptive sound field compensation, characterized in that, include: The environmental monitoring module is used to acquire the corresponding environmental noise signal from the corresponding environmental terminal within the portable hearing monitoring device, and to preprocess the environmental noise signal through the environmental terminal. The hearing monitoring module is used to acquire corresponding audio test signals and user feedback information from the corresponding test detection end within the portable hearing monitoring device; The data transmission module is used to set the environmental transmission channel protocol and the test transmission channel protocol corresponding to the data processing end in the portable hearing monitoring device, and to transmit the corresponding data information according to the corresponding transmission channel protocol. The data transmission module includes: The transmission management unit is used to obtain the classification intervals corresponding to the environment end and the detection end, set the corresponding environment classification interval protocol and test classification interval protocol according to the corresponding feature extraction results in each classification interval, sort the environment classification interval protocol and test classification interval protocol respectively, and construct the transmission channel protocol architecture on the sorting results. Based on the architecture results, the corresponding environment transmission channel protocol and test transmission channel protocol are generated respectively. The distribution of microphone arrays in the environment and detection ends is obtained respectively. Based on the distribution of microphone arrays, corresponding environmental transmission sub-channels and test transmission sub-channels are set. The environmental transmission channel protocol and test transmission channel protocol are mapped to each environmental transmission sub-channel and test transmission sub-channel respectively. The data transmission unit is used for the environmental end and the detection end to transmit the corresponding data information according to the environmental transmission sub-channel and the test transmission sub-channel connected to the data processing end, respectively, to obtain the matching analysis results of the corresponding data information with the classification interval protocol in sequence during the transmission process, and to generate classification matching codes based on the matching results; The environmental compensation module is used to set an environmental sequence frame based on the acquisition time according to the environmental noise signal obtained within the environmental transmission channel protocol, obtain the initial sound field environmental parameters based on the corresponding environmental noise signal within the environmental sequence frame, perform environmental adaptive compensation on the audio test signal based on the initial sound field environmental parameters, and obtain the initial audio test signal. The test compensation module is used to compare and analyze the obtained audio test signal with the initial audio test signal to obtain deviation audio test information, and to adaptively compensate the initial audio test signal based on the deviation audio test information to obtain a comprehensive audio test signal. The detection and analysis module is used to obtain user feedback information corresponding to the comprehensive audio test signal, extract features from the user feedback information, construct a hearing test comparison chart, analyze and process the hearing test comparison chart, and obtain the corresponding hearing test data.
2. The portable precision hearing detection system based on adaptive sound field compensation according to claim 1, characterized in that, The environmental monitoring module includes: The environmental monitoring and processing unit is used to acquire corresponding environmental noise signals from the corresponding microphone array at the environmental end, mark the environmental noise signals according to the corresponding microphone number and acquisition time, and decompose the marked environmental noise signals based on the independent component analysis method to obtain the corresponding independent source signals. The environmental monitoring classification unit is used to extract features from historical source signals, obtain corresponding time-domain and frequency characteristics, and set corresponding classification intervals based on the feature extraction results.
3. The portable precision hearing detection system based on adaptive sound field compensation according to claim 2, characterized in that, The hearing monitoring module includes: The test monitoring and processing unit is used to acquire audio test signals from the ear of the tester, mark the acquired audio test signals according to the acquisition time, extract features from the audio test signals to obtain audio test feature data, and classify the audio test feature data. The feedback monitoring unit is used by the detection end to acquire user feedback information from the tester during the test, and to mark the user feedback information according to the collection time.
4. A portable, precise hearing detection system based on adaptive sound field compensation according to claim 3, characterized in that, The environmental compensation module includes: The data preprocessing unit is used to obtain the historical environmental noise signals and historical sound field environmental parameters corresponding to the corresponding classification intervals, set the corresponding classification noise sound field datasets, and analyze and train the obtained classification noise sound field datasets based on deep learning algorithms to construct classification sound field environmental parameter models. The model pre-matching unit is used to acquire the environmental noise signal transmitted from the environmental end and the corresponding classification matching code, acquire the corresponding classification sound field environmental parameter model according to the corresponding classification matching code, and analyze and process the acquired environmental noise signal through the corresponding classification sound field environmental parameter model to obtain the corresponding channel sound field environmental parameters. The fusion processing unit is used to set up corresponding three-dimensional channel space models according to the environmental transmission sub-channels; to map the environmental noise signals and channel sound field environmental parameters obtained by the corresponding environmental transmission sub-channels into the corresponding three-dimensional channel space models for visualization processing, to obtain the corresponding three-dimensional environmental space images, and to set up sequence frames according to the acquisition time, to fuse the environmental noise signals and channel sound field environmental parameters corresponding to the corresponding pixels in each sequence frame corresponding to the same acquisition time, to obtain the initial sound field environmental parameters corresponding to the acquisition time. An environmental compensation unit is used to acquire the initial audio test signal.
5. A portable, precise hearing detection system based on adaptive sound field compensation according to claim 4, characterized in that, The process by which the environmental compensation unit acquires the initial audio test signal includes: The initial sound field environment parameters and corresponding audio test signals corresponding to the acquisition time are obtained. An adaptive algorithm is selected based on the initial sound field environment parameters. The adaptive algorithm includes LMS, FDAF and beamforming algorithm. The environmental interference is iteratively compensated for based on the selected adaptive algorithm. The audio test signals are then adapted based on the compensated parameters to generate the initial audio test signals.
6. A portable, precise hearing detection system based on adaptive sound field compensation according to claim 5, characterized in that, The test compensation module includes: The audio comparison unit is used to obtain the classification interval corresponding to the initial audio test signal, as well as the classification matching code and corresponding test monitoring signal obtained by the test transmission sub-channel. The corresponding classification matching code is compared and analyzed with the corresponding classification interval in turn to determine whether they belong to the same classification interval. The initial audio test signal and test monitoring information that do not belong to the same classification interval are compared and analyzed first to obtain the corresponding deviation audio test signal. The test fusion unit is used to set up corresponding three-dimensional test space models according to the test transmission sub-channels, map the test monitoring signals and deviation audio test signals obtained by the corresponding test transmission sub-channels into the three-dimensional test space models, obtain the corresponding three-dimensional test space images, and perform fusion processing on the deviation audio test signals at corresponding pixels in the three-dimensional test space images corresponding to the same acquisition time to obtain the deviation fusion signal corresponding to the acquisition time. The test compensation unit is used to select the appropriate adaptive algorithm based on the obtained deviation fusion signal. The appropriate adaptive algorithm performs adaptive compensation on the initial audio test signal based on the deviation fusion signal to obtain the comprehensive audio test signal.
7. A portable, precise hearing detection system based on adaptive sound field compensation according to claim 6, characterized in that, The detection and analysis module includes: The image comparison unit is used to acquire the corresponding comprehensive audio test signal and user feedback information, and to compare and mark the comprehensive audio test signal with the user feedback information. The user feedback information includes subjective feedback information and otoacoustic reflex signal. Feature extraction is performed on the user feedback information to obtain feedback feature signals. The obtained feedback feature signals are used to establish hearing test comparison charts corresponding to the comprehensive audio test signal based on the corresponding time information. The detection and analysis unit is used to analyze and process the hearing test comparison charts corresponding to each feedback feature signal, obtain the corresponding hearing test feature data, and perform comprehensive analysis on the obtained hearing test feature data to determine whether there are any contradictory hearing test feature data. If there are no contradictory hearing test feature data, the corresponding hearing test data is generated based on the comprehensive analysis results.