A tinnitus intervention method, device and equipment based on a brain-computer interface

Abstract

The application relates to a tinnitus intervention method, device and equipment based on a brain-computer interface, relates to the technical field of tinnitus digital intervention, and comprises the following steps: processing pure tone audiometry results of a target user by using a matching method to obtain input parameters such as tinnitus frequency and loudness. Real-time dynamic processing of original audio is performed in a buffer area to generate an audio block, band-pass filtering is performed in combination with the tinnitus input parameters, and a band-pass / band-stop filter is constructed. Key noise is introduced by using random noise, the noise is optimally superimposed on the audio block, signal gain is analyzed, notch enhancement is performed based on a notch / channel mode according to the band-stop filter and the gain, and notch music is generated. Finally, the notch music is combined with neural feedback information based on the brain-computer interface, comprehensive tinnitus intervention is performed, the notch music is adjusted, and the result of the tinnitus intervention is determined. It can be seen that the brain-computer interface and the notch music therapy are combined to realize the complementary advantages, and the tinnitus digital intervention effect is improved.

Description

Technical Field

[0001] This application relates to the field of digital intervention technology for tinnitus, and in particular to a tinnitus intervention method, device, and equipment based on a brain-computer interface. Background Technology

[0002] Traditional digital interventions for tinnitus primarily involve sound therapy, including masking therapy and notched music therapy. Masking therapy reduces the user's perception of tinnitus by playing white noise or user-tuned sounds. Notched music therapy uses generated notched music to improve tinnitus through lateral inhibition (a neural mechanism). The specific process is as follows: First, pure-tone audiometry is performed to determine the tinnitus frequency; notch filtering is applied to the music in the tinnitus frequency band (generally around 1000 Hz); gain is then applied to the edges of the notch-filtered frequency band.

[0003] However, existing tinnitus intervention methods using notch music can only input complete audio files and output complete processed audio during notch music processing. Furthermore, the entire notch processing process can only rely on fixed parameters, which does not conform to the user's actual tinnitus condition, resulting in poor digital intervention effects for tinnitus. Summary of the Invention

[0004] This application provides a brain-computer interface-based tinnitus intervention method, device, and equipment. By integrating brain-computer interface, cognitive behavioral therapy, and notch music therapy, a digital and interactive comprehensive tinnitus intervention program is formed. On the one hand, through streaming processing, the processing and playback of notch music are synchronized, improving the smoothness of interactive operation. On the other hand, by adopting multiple technologies and digitally innovating the tinnitus intervention program, the tinnitus intervention method is made more suitable for the user's actual situation, breaking through the scenario limitations of traditional solutions, and providing a more flexible and accessible solution for tinnitus intervention. This improves the effect of digital tinnitus intervention and solves the problems of untimely audio processing and poor tinnitus intervention effect in existing technologies for digital tinnitus intervention.

[0005] Firstly, this application provides a tinnitus intervention method based on a brain-computer interface, including:

[0006] Obtain audiograms and hearing types generated from digital pure-tone audiometry of target users, including groups with auditory perception abnormalities requiring intervention;

[0007] Based on audiograms and hearing types, tinnitus frequency and tinnitus loudness are measured for target users using a matching method to obtain tinnitus input parameters. The tinnitus input parameters include the tinnitus center frequency obtained from the tinnitus frequency measurement and the relative tinnitus loudness obtained from the tinnitus loudness measurement.

[0008] Within the established music buffer, real-time or near-real-time processing is performed on the raw audio data. Based on the initialization parameters constructed from the tinnitus input parameters, and combined with the pre-processed audio blocks represented by a two-dimensional matrix, bandpass filtering is performed to construct bandpass and bandstop filters.

[0009] For the audio block, random noise and a bandpass filter are used to introduce key target noise, and the target noise is superimposed with the audio block signal using the optimal gain to obtain gain information;

[0010] Based on the cutting mode and channel mode selected by the target user, real-time processing and cutting enhancement are performed according to the band-stop filter and gain information to generate an audio signal that meets the requirements of acoustic intervention, which is then used as cutting music.

[0011] By monitoring the target user's neural activity in real time through a brain-computer interface, the notch music is combined with neural feedback information, allowing the notch music to be adjusted based on the neural feedback information. At the end of the tinnitus intervention, the training results of the current tinnitus intervention training are obtained.

[0012] Optionally, based on audiograms and hearing types, a matching method is used to measure the tinnitus frequency and loudness of the target user to obtain tinnitus input parameters, including:

[0013] Based on the frequency range, the frequency coordinate system is divided into average laws to construct a frequency adjustment benchmark;

[0014] When the target user conducts a pure tone test, based on the frequency adjustment benchmark, the pitch adjustment information of the target user through interactive operation is determined, and the frequency of the pitch adjustment information is taken as the tinnitus frequency corresponding to the user's perception of tinnitus sound.

[0015] The matching method is used to compare the tinnitus frequency with pure tones of known frequencies to determine the tinnitus center frequency;

[0016] Identify the volume adjustment information of the target user through interactive operations, and determine the tinnitus volume to mask the tinnitus sound based on the volume adjustment information;

[0017] The relative tinnitus loudness is determined by comparing the tinnitus volume with a volume baseline percentage using a matching method.

[0018] Optionally, within the established music buffer, real-time or near-real-time processing is performed on the raw audio data. Based on initialization parameters constructed from tinnitus input parameters, and combined with pre-processed audio blocks represented by two-dimensional matrices, bandpass filtering is performed to construct bandpass and bandstop filters, including:

[0019] The target user's preset tinnitus side parameters and processing mode, combined with the tinnitus input parameters, are used as user setting parameters. The obtained algorithm initialization parameters and user setting parameters are used as the initialization parameters for the notch music generation algorithm.

[0020] According to the preset time segments, the raw audio data currently being played is preprocessed in real time or near real time to obtain audio blocks represented by a two-dimensional matrix;

[0021] Using initialization parameters and audio blocks as input, the core frequency parameters are determined through filtering and bandpass processing. The core frequency parameters include at least the reference frequency of the notch, the center frequency, the bandpass filter width, and the bandstop filter width.

[0022] Based on the center frequency and the width of the bandpass filter, a bandpass filter is constructed using a preset filter; and based on the center frequency and the width of the bandstop filter, a bandstop filter is constructed using a preset filter.

[0023] Before processing the raw audio data, a music buffer is constructed, and the raw audio data is loaded into a memory matrix and cached. During the processing of the raw audio data, each piece of data processed is an audio block in the raw audio data, and each audio block is determined by the segment index and segment size.

[0024] Optionally, the core frequency parameters are determined through bandpass filtering, using initialization parameters and audio blocks as input, including:

[0025] Taking the notch audio range in the initialization parameters as input, according to Reference frequency for calculating the gain range of the notch edge ;

[0026] according to Calculate the center frequency of the center position of the notch band. ;

[0027] according to The coverage of the notch frequency band and edge gain region is determined to obtain the bandpass filter width. ;

[0028] according to Calculate the width of the band-stop filter It is used to accurately locate the core frequency band of the cut that needs to be attenuated.

[0029] Optionally, for the audio block, random noise and a bandpass filter are used to introduce key target noise, and the target noise is superimposed on the audio block signal using optimal gain to obtain gain information, including:

[0030] For an audio block, random noise is generated and processed by a bandpass filter to obtain the key target noise, which contains frequency components that retain only the frequency bands at the edges of the notch.

[0031] Based on the edge gain in the initialization parameters, the optimal gain is determined according to the debugging information of the target user in the visualized spectrum;

[0032] The target noise is superimposed with the original signal corresponding to the audio block using the optimal gain to obtain gain information, which is used to provide energy compensation for the cut edge.

[0033] Optionally, based on the chipping mode and channel mode selected by the target user, real-time processing and chipping enhancement are performed according to the band-stop filter and gain information to generate an audio signal that meets the acoustic intervention requirements, as chipped music, including:

[0034] Based on the cleaving pattern selected by the target user, real-time processing and cleaving enhancement are performed according to the band-stop filter and gain information to generate cleaving music that meets the requirements of acoustic intervention.

[0035] Based on the channel mode selected by the target user, output cut music in the corresponding channel;

[0036] During the output of the cut music, after the cut music of the current audio block finishes playing, the next audio block is processed immediately, until the cut music of all audio blocks has finished playing.

[0037] Optionally, the neural activity state of the target user can be monitored in real time via a brain-computer interface. The notch music can be combined with neural feedback information, allowing the notch music to be adjusted based on this feedback. At the end of the tinnitus intervention, the training results of the current tinnitus intervention training can be obtained, including:

[0038] Replace the background music used in the neurofeedback scenario with notch music, and use a brain-computer interface to monitor the target user's neural activity state in real time to obtain neurofeedback information;

[0039] Based on the neural feedback information, the final volume of the notch music output is adjusted. At the end of the tinnitus intervention, the practice data for the current tinnitus intervention cycle is determined based on the tinnitus loudness measurement information of the target user.

[0040] Based on the practice data of the current cycle, combined with the practice data of the previous cycle, the training results of tinnitus intervention training are determined.

[0041] Optionally, obtain the audiogram and hearing type generated from the digital pure-tone audiometry of the target user, including:

[0042] During the digital pure-tone audiometry process for target users, the pure tone is output from low to high frequencies through air conduction audiometry.

[0043] According to the threshold determination algorithm, the air conduction threshold is determined based on the feedback operation of the target user during pure tone output.

[0044] The air conduction threshold is marked on the preset pure tone audiogram to generate an audiogram. Hearing loss is then analyzed based on the air conduction threshold to determine the hearing type.

[0045] Secondly, this application provides a tinnitus intervention device based on a brain-computer interface, comprising:

[0046] The pure-tone audiometry module is used to perform the steps of generating audiograms and audio types.

[0047] The frequency and loudness measurement module is used to perform the steps of measuring tinnitus frequency and tinnitus loudness.

[0048] The preprocessing module is used to perform real-time or near-real-time processing on the raw audio data;

[0049] The bandpass filtering module is used to perform the bandpass filtering process on the audio block;

[0050] The noise processing module is used to perform the step of superimposing the target noise and the audio block into signals using optimal gain.

[0051] The track-cutting music generation and output module is used to perform track-cutting processing and output based on the track-cutting mode and channel mode selected by the target user;

[0052] The tinnitus intervention training result determination module is used to perform the steps of real-time monitoring of the target user's neural activity state and adjusting the notch music based on neural feedback information.

[0053] Thirdly, this application provides an electronic device, which is an electronic device with data processing, storage and communication functions, and is used to implement the steps of the tinnitus intervention method based on brain-computer interface as described in any one of the first aspects.

[0054] In summary, this application first utilizes pure-tone audiometry to test the target user's audiogram and hearing type. Then, it uses a matching method to determine tinnitus input parameters such as the tinnitus center frequency and tinnitus loudness. Within a buffer, the original audio is processed in real time to generate audio blocks. Bandpass filtering is then applied to these blocks in conjunction with the tinnitus input parameters to construct a bandpass / bandstop filter. Random noise and a bandpass filter are used to introduce key target noise. This target noise is then optimally superimposed on the audio blocks, and the signal gain is analyzed to obtain gain information. Next, based on the user-selected notch / channel mode, notch enhancement is performed according to the bandstop filter and gain information, generating and outputting notch music corresponding to the audio blocks. Finally, the target user's neural activity is monitored in real time via a brain-computer interface. The notch music is combined with neural feedback for comprehensive tinnitus intervention. The notch music is adjusted based on the neural feedback information, and the tinnitus intervention result is obtained when the whispered intervention ends. Therefore, this application integrates a brain-computer interface and notch music therapy to form a digital, interactive tinnitus intervention program. By processing the music in real time, the processing and playback of the music can be synchronized, improving the smoothness of the interactive operation. On the other hand, by innovating the tinnitus intervention solution digitally, the tinnitus intervention is more in line with the actual situation of the user, improving the effect of digital tinnitus intervention and solving the problems of untimely audio processing and poor tinnitus intervention effect in the existing technology in digital tinnitus intervention. Attached Figure Description

[0055] The accompanying drawings, which are incorporated in and form part of this specification, illustrate embodiments consistent with this application and, together with the description, serve to explain the principles of this application.

[0056] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, for those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0057] Figure 1 A flowchart illustrating a brain-computer interface-based tinnitus intervention method provided in this application embodiment;

[0058] Figure 2 This is a flowchart illustrating the steps of a brain-computer interface-based tinnitus intervention method provided in an optional embodiment of this application.

[0059] Figure 3 A structural block diagram of a tinnitus intervention method device based on a brain-computer interface provided in this application embodiment;

[0060] Figure 4 This is an optional example of an electronic device provided in this application. Detailed Implementation

[0061] To make the objectives, technical solutions, and advantages of the embodiments of this application clearer, the technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0062] To facilitate understanding of the embodiments of this application, further explanations and descriptions will be provided below in conjunction with the accompanying drawings and specific embodiments. These embodiments do not constitute a limitation on the embodiments of this application.

[0063] Figure 1 This is a flowchart illustrating a brain-computer interface-based tinnitus intervention method provided in an embodiment of this application. The brain-computer interface-based tinnitus intervention method may specifically include the following steps:

[0064] Step 110: Obtain the audiogram and hearing type generated by the target user's digital pure tone audiometry.

[0065] The target users include those with auditory perception abnormalities who require intervention.

[0066] In the relevant technologies, the existing tinnitus intervention methods using notch music therapy use pure tone audiometry parameters when generating notch music. This pure tone audiometry process generally requires the subject to be in a hospital under the guidance of an audiologist, which has obvious limitations in application scenarios and cannot be adapted to the personalized treatment scenarios of tinnitus patients.

[0067] This embodiment improves the pure-tone audiometry process by digitizing the existing pure-tone audiometry process, enabling users to perform measurements via mobile phone and headphones. The process mainly includes: the target user (hereinafter referred to as "subject" or simply "user") wears headphones (which can be wired or wireless headphones, mainly connected to terminals such as mobile phones and tablets) and performs pure-tone audiometry by playing pure tones through the mobile phone. During the audiometry, the sound is tested sequentially from low to high frequencies. By monitoring the subject's operation, frequency thresholds are measured, and the thresholds are marked on the pure-tone audiogram to form an audiogram. Based on the threshold analysis and comparison of high and low frequency thresholds, the degree of hearing loss (such as high-frequency hearing loss or low-frequency hearing loss) is determined, and the hearing type is obtained.

[0068] It should be noted that the tinnitus intervention method provided in this embodiment can offer digital tinnitus intervention treatment to groups with auditory perception abnormalities requiring intervention. These groups mainly include tinnitus patients, as well as those experiencing conditions such as hyperacusis or auditory hallucinations. This embodiment does not limit the specific groups targeted for intervention.

[0069] Step 120: Based on the audiogram and hearing type, the tinnitus frequency and loudness of the target user are measured using a matching method to obtain the tinnitus input parameters.

[0070] The tinnitus input parameters include the tinnitus center frequency obtained from tinnitus frequency measurement and the relative tinnitus loudness obtained from tinnitus loudness measurement.

[0071] In practice, a matching method is used to measure the tinnitus frequency and loudness of the subject. Specifically, measuring tinnitus frequency involves comparing the frequency of the tinnitus sound perceived by the subject with pure tones of known frequencies to determine the tinnitus frequency, i.e., the tinnitus center frequency. Measuring tinnitus loudness involves adjusting the volume of the measured tinnitus frequency, allowing the user to determine the volume that just masks their tinnitus sound, thus determining the tinnitus loudness, i.e., the relative tinnitus loudness.

[0072] Step 130: Within the established music buffer, perform real-time or near-real-time processing based on the original audio data, and combine the initialization parameters constructed based on the tinnitus input parameters with the pre-processed audio blocks represented by a two-dimensional matrix to perform bandpass filtering processing, thereby constructing a bandpass filter and a bandstop filter.

[0073] In this embodiment, unlike existing track-cutting music processing schemes, this embodiment uses a real-time or near-real-time approach to perform track-cutting processing on the music. Real-time processing includes, but is not limited to, various real-time processing methods such as streaming processing, batch processing, and segmented processing. Preferably, streaming processing can be used to process track-cut music. Streaming processing establishes an audio buffer, enabling processing and playback simultaneously, thereby significantly reducing the waiting time for track-cutting processing and supporting real-time adjustment of track-cutting parameters.

[0074] This embodiment introduces a notched music generation algorithm for generating notched music. This algorithm takes the input music file as the raw audio data and incorporates tinnitus frequency and edge gain to achieve the final notched music generation.

[0075] In this implementation, an audio buffer can be created, and a copy of the input audio data, `audioDataOutput`, can be created. The original audio is loaded into a memory matrix and cached, avoiding direct modification of the original data and ensuring the independence of the processing. The notching processing system adopts a streaming audio processing method. Its input data is the original audio data. During each preprocessing, a certain audio block of the original audio data is used as the processing target to generate the current audio block, which is represented in the form of a two-dimensional matrix. Before notching processing by the algorithm, initialization parameters are constructed based on the tinnitus input parameters. The initialization parameters mainly consist of two types of parameters: user-set parameters and algorithm internal initialization parameters. The user-set parameters mainly include the measured tinnitus input parameters and preset edge gains, etc. Then, the initialization parameters are introduced, and the core frequency parameters (also known as filter parameter calculation) are calculated using the bandpass filtering algorithm based on the initialization parameters and the audio block. Finally, using the calculated core frequency parameters, the filter is instantiated and configured to construct the bandpass filter and the bandstop filter.

[0076] Step 140: For the audio block, key target noise is introduced using random noise and a bandpass filter, and the target noise is superimposed on the audio block signal using optimal gain to obtain gain information.

[0077] To enhance the frequency band energy at the notch edge, this embodiment introduces random noise and removes irrelevant frequency bands of the random noise through bandpass filtering, resulting in key target noise with a gain effect. The random noise in this embodiment can be, for example, Gaussian noise, white noise, or pink noise. After bandpass filtering, the frequency band at the notch edge is preserved.

[0078] Subsequently, based on the optimal gain set by the locked user, the target noise and the original signal corresponding to the audio block are superimposed with the optimal gain to provide energy compensation for the cut edge and obtain the gain information.

[0079] Step 150: Based on the cutting mode and channel mode selected by the target user, perform real-time processing and cutting enhancement according to the band-stop filter and gain information to generate an audio signal that meets the requirements of acoustic intervention, as cutting music.

[0080] This embodiment implements multi-mode audio notching processing, supporting personalized notching processing in various modes. Specifically, it involves notching modes and channel modes, both of which can be preset by the target user. This embodiment does not limit the specific notching and channel modes. For example, notching modes may include, but are not limited to, mono notching, binaural notching, local frequency band notching, and no notching (i.e., no processing of the original audio data, mainly outputting the original audio); channel modes may include, but are not limited to, mono, stereo, and multi-channel (such as multi-channel directional notching).

[0081] In the specific implementation, the subject can choose different modes according to their needs. For example, if the subject has tinnitus in one ear canal, they can choose mono-ear notching, mono-channel, etc. This embodiment does not impose any restrictions on this. For example, each time notching is performed, firstly, the channel mode selected by the subject is determined. This channel can be understood as the channel on the side where the subject has tinnitus, and only this channel is retained after notching. Then, according to the notching mode selected by the subject, it is determined whether to perform mono-ear notching, binaural notching, or no notching, etc., and notching is performed on the channel on the side where the subject has tinnitus. Then, notching is performed. Notching is mainly achieved by real-time processing and notching enhancement of the audio block by a band-stop filter and gain information to achieve the notching effect. This process mainly includes: ① superimposing gain information on the original audio data to enhance the frequency band energy at the edge of the notch; ② attenuating the audio signal of the core notch frequency band through a band-stop filter.

[0082] Through groove processing, this embodiment generates grooved music that meets the requirements of acoustic intervention, and the grooved music corresponds to the processed audio segment.

[0083] In this embodiment, after generating the notch music, the notch music can be output, such as by playing it through headphones or other audio devices, to provide digital intervention for tinnitus in the subjects.

[0084] It should be noted that the core principle of notch processing in different modes in this embodiment is similar to that in the above example, and will not be repeated here. Specifically, in single-ear notching, the original audio is retained and not processed on the non-tinnitus side not selected by the subject, and the original audio is output on the non-tinnitus side when outputting notched music.

[0085] Step 160: Monitor the target user's neural activity state in real time through a brain-computer interface, combine notch music with neural feedback information, and adjust the notch music based on neural feedback information. At the end of the tinnitus intervention, obtain the training result of the current tinnitus intervention training.

[0086] In its specific implementation, this embodiment combines notched music with neurofeedback scenarios to achieve digital tinnitus intervention. Neurofeedback scenarios may include, but are not limited to, meditation scenarios, attention training, relaxation training, and biofeedback.

[0087] Taking a meditation scenario as an example, when the target user meditates in a meditation practice setting, the background music is replaced with notched music. Then, the target user's EEG signals are monitored via a brain-computer interface to obtain neurofeedback information. This neurofeedback information serves as the meditation information, and a meditation score is determined based on this information. The final volume, i.e., the volume of the notched music, is adjusted accordingly. For example, a higher meditation score corresponds to a louder background volume, which better masks tinnitus, making the user feel more relaxed and comfortable. Finally, at the end of the meditation (i.e., the end of the tinnitus intervention), the subject can measure the tinnitus loudness and compare it to historical loudness levels to assess the practice effectiveness score, thereby evaluating the training results of this tinnitus intervention.

[0088] As can be seen, this application's embodiments, by combining brain-computer interfaces and notch music therapy, form a digital, interactive comprehensive tinnitus intervention program, achieving a real-time audio processing mode: employing streaming buffer technology, it changes the traditional waiting mode of complete file processing, enabling synchronous processing and playback of notch music, significantly shortening the response time after user parameter adjustments and improving the smoothness of interactive operations. Secondly, it enhances intervention adaptability through multi-therapy synergy. Specifically, through an integrated intervention mechanism, it combines the acoustic intervention of notch music with neurofeedback information from the brain-computer interface. For example, in a meditation scenario, it dynamically links the meditation score from neurofeedback with the volume adjustment of notch music, achieving complementary therapeutic advantages and making the intervention program more aligned with individual neurofeedback characteristics and tinnitus acoustic needs. The introduction of real-time parameter adjustment functionality allows users to independently adjust the notch frequency and edge gain and listen instantly, overcoming the limitations of fixed parameters in traditional programs and improving the personalized adaptability of the intervention program. This improves the effectiveness of digital tinnitus intervention and solves problems such as untimely audio processing and poor tinnitus intervention effects in existing digital tinnitus intervention technologies.

[0089] Reference Figure 2 The diagram illustrates a step-by-step flowchart of a brain-computer interface-based tinnitus intervention method according to an optional embodiment of this application. The method may specifically include the following steps:

[0090] Step 210: Obtain the audiogram and hearing type generated by the target user's digital pure tone audiometry.

[0091] The target users include those with auditory perception abnormalities who require intervention.

[0092] Optionally, this embodiment may acquire the audiogram and hearing type generated by the target user's digital pure-tone audiometry, which may include: during the target user's digital pure-tone audiometry, outputting the pure tone from low to high frequencies through air conduction audiometry; determining the air conduction threshold based on the target user's feedback during the pure-tone output process according to a threshold determination algorithm; marking the air conduction threshold on a preset pure-tone audiogram to generate an audiogram, and analyzing the hearing loss based on the air conduction threshold to determine the hearing type.

[0093] In this embodiment, the digital pure-tone audiometry process can be implemented through the following steps:

[0094] Preparation before inspection:

[0095] ① Environment and Equipment: Remind users to conduct the test in a quiet place, measure the ambient noise through the terminal device's microphone, and ensure that the ambient noise is ≤30 decibels; recommend that users wear headphones; check and record the headphone connection status; if headphones are not worn, prompt the user to place the device flat on the table and repeat the measurement 50 times; if the headphone connection status is inconsistent with the previous measurement in the next measurement, trigger a prompt, indicating that the two measurements used different output devices and the measurement results are not comparable;

[0096] ②Subject preparation:

[0097] Tip: Check the external auditory canal: remove cerumen (earwax) to avoid external ear blockage affecting the results.

[0098] Explain the process: Inform users to click the button based on whether they can hear a sound: "Can hear", "Cannot hear".

[0099] Air conduction audiometry:

[0100] ① Listening test order: test the right ear first, then the left ear, and test from low frequency (e.g., 125Hz) to high frequency (e.g., 8000Hz) in sequence.

[0101] ②Threshold determination:

[0102] Sound delivery method: Start with an intensity slightly higher than the estimated hearing threshold (e.g., 40% volume). If the subject can hear it, gradually reduce the intensity (10% volume each time) until they cannot hear it. Then increase the intensity in 5% increments until the subject can just hear it. Repeat twice to confirm the threshold (the minimum intensity at which the subject can hear 50% of the sounds).

[0103] Frequency points: routine tests at 125, 250, 500, 1000, 2000, 4000, and 8000 Hz.

[0104] Results Recording and Presentation

[0105] ① Draw an audiogram: Mark the air conduction threshold on the pure tone audiogram (the horizontal axis is frequency and the vertical axis is volume percentage).

[0106] ② Determine the type of hearing loss: If the high-frequency hearing threshold is higher than the low-frequency hearing threshold, then it is high-frequency hearing loss.

[0107] The above steps can systematically assess the degree and nature of hearing loss, providing a basis for diagnosis and intervention.

[0108] Step 220: Based on the audiogram and hearing type, the tinnitus frequency and loudness of the target user are measured using a matching method to obtain the tinnitus input parameters.

[0109] The tinnitus input parameters include the tinnitus center frequency obtained from tinnitus frequency measurement and the relative tinnitus loudness obtained from tinnitus loudness measurement.

[0110] Optionally, the above-mentioned method, based on audiograms and hearing types, uses a matching method to measure the tinnitus frequency and loudness of the target user to obtain tinnitus input parameters. Specifically, this may include: dividing the frequency coordinate system into mean laws according to the frequency range to construct a frequency adjustment benchmark; when the target user performs a pure tone test, determining the pitch adjustment information of the target user's frequency adjustment through interactive operation based on the frequency adjustment benchmark, and taking the frequency of the pitch adjustment information as the tinnitus frequency corresponding to the user's perception of the tinnitus sound; using a matching method to compare the tinnitus frequency with pure tones of known frequencies to determine the tinnitus center frequency; determining the volume adjustment information of the target user's volume adjustment through interactive operation, and determining the tinnitus volume to mask the tinnitus sound based on the volume adjustment information; and using a matching method to compare the tinnitus volume with the volume benchmark percentage to determine the relative tinnitus loudness.

[0111] In the specific implementation, for the measurement of the tinnitus center frequency, this embodiment uses the matching method to measure the tinnitus frequency by comparing the tinnitus sound perceived by the patient with a pure tone of known frequency, thereby determining the tinnitus center frequency.

[0112] To achieve frequency adjustment, subjects can adjust the frequency through interactive operations (such as sliding adjustment, button selection, etc.). Taking sliding adjustment as an example, the terminal device used by the subject, such as a mobile phone, can be equipped with a slider for frequency adjustment (it can be a physical slider or a virtual slider implemented through the touch screen; this embodiment does not limit this, and the sliding method can be left and right sliding or up and down sliding). Subjects can adjust the frequency by sliding the slider left and right.

[0113] In this embodiment, frequency adjustment can be performed according to a frequency adjustment benchmark. Specifically, firstly, a frequency range is defined, which refers to dividing the frequency coordinate system according to the twelve-tone equal temperament to form a frequency adjustment benchmark. Then, each time the subject performs frequency adjustment, it is adjusted according to the frequency adjustment benchmark. For example, if the frequency adjustment benchmark divides the frequency range into 12 levels from low to high, then each frequency adjustment increases the frequency by one level.

[0114] When the subject perceives the tinnitus sound, the frequency adjustment operation is stopped. At this point, the frequency corresponding to the pitch information is adjusted, which is the tinnitus frequency perceived by the subject. Finally, the tinnitus frequency is compared with pure tones of known frequencies using a matching method to determine the tinnitus center frequency.

[0115] For the measurement of relative tinnitus loudness, the matching method is used to determine the relative tinnitus loudness by adjusting the volume of the measured tinnitus frequency and having the user determine the volume that just masks the tinnitus sound.

[0116] In tinnitus loudness measurement, subjects can also adjust the volume through interactive operations. The interactive operations can be defined according to needs, including but not limited to: sliding adjustment and button selection through physical buttons or touch, as well as voice adjustment through voice.

[0117] Taking the sliding adjustment as an example, the mobile phone or other terminal device used by the subject can be equipped with a volume slider for volume adjustment (its appearance / usage can be referred to the frequency adjustment slider mentioned above, which will not be repeated in this embodiment). The subject can adjust the volume by sliding the volume slider, where the volume changes in 1% increments. The volume percentage corresponding to where the volume slider finally stops (i.e., the volume adjustment information) is the relative tinnitus loudness.

[0118] During the tinnitus loudness measurement process, the subject determines the volume that just masks their tinnitus sound, and the corresponding percentage is the relative tinnitus loudness.

[0119] The environment and equipment for tinnitus center frequency measurement can be consistent with those for pure-tone audiometry. For tinnitus center frequency measurement, subject preparation includes: explaining the procedure: "Next, slide the slider. You will hear sounds of different pitches. Slide the slider to the position closest to the pitch of your tinnitus." Confirm that the patient understands the concept of "frequency" (pitch) to avoid confusion with loudness (volume). For relative tinnitus loudness measurement, user preparation includes: explaining the procedure: "Next, slide the slider. You will hear sounds of different volumes. Slide the slider to the position that just masks your tinnitus sound." Confirm that the patient understands the concept of "frequency" (pitch) to avoid confusion with loudness (volume).

[0120] As can be seen, this embodiment improves the convenience and accuracy of digital audiometry. In the home-based self-service audiometry experience: the professional pure-tone audiometry process from hospitals is digitized to a mobile device. Through functions such as environmental noise monitoring and headphone status prompts, the operation process is simplified, allowing users to complete full-frequency hearing assessments without relying on professional facilities or personnel, thus expanding the accessibility of audiometry scenarios. In the refined tinnitus parameter measurement: by using a matching method to adjust frequency and loudness with fine-grained step settings, the tinnitus characteristics are quantitatively captured, providing more accurate basic data support for subsequent intervention plans.

[0121] Step 230: The target user's preset tinnitus side parameters and processing mode, combined with the tinnitus input parameters, are used as user-set parameters. The obtained algorithm initialization parameters and user-set parameters are used as the initialization parameters for the notch music generation algorithm.

[0122] Step 240: According to the preset time segment, the currently playing raw audio data is preprocessed in real time or near real time to obtain audio blocks represented by a two-dimensional matrix.

[0123] Before processing the raw audio data, a music buffer is constructed, and the raw audio data is loaded into a memory matrix and cached. During the processing of the raw audio data, each piece of data processed is an audio block in the raw audio data, and each audio block is determined by the segment index and segment size.

[0124] Steps 230-240 are described uniformly as follows:

[0125] In this embodiment, user-configurable parameters include tinnitus center frequency, edge gain, tinnitus side, and processing mode (Type). Algorithm initialization parameters refer to the internal initialization parameters of the algorithm, mainly including the notch frequency range (low frequency lFreq, high frequency hFreq), edge gain edgeGain, dual-channel audio data audioDataInput, sampling rate audioSampleRate, chunk length (chunkSeconds), and temporary path configuration for audio segments. This embodiment can also verify the initialization parameters, primarily verifying the validity of the notch frequency range: ensuring 1000 < lFreq < hFreq to avoid invalid parameters causing processing abnormalities.

[0126] In practical implementation, a groove processing system can be built to process grooved music.

[0127] For processing raw audio data using streaming audio processing, firstly, a copy of the input audio data, named `audioDataOutput`, is created. The raw audio is loaded into a memory matrix and cached to avoid directly modifying the original data and ensure the independence of the processing. Then, the segmentation processing system processes a specific chunk of the raw audio data each time, determined primarily by the `chunkIndex` and `chunkSize`. The system reads data in time segments (e.g., every 2 seconds) and forms a two-dimensional matrix as the current processing block, i.e., the audio block. By establishing an audio buffer through streaming processing, processing and playback can proceed simultaneously, significantly reducing the waiting time for segmentation processing.

[0128] Step 250: Using initialization parameters and audio blocks as input, determine the core frequency parameters through filtering and bandpass processing.

[0129] The core frequency parameters include at least the reference frequency of the slit, the center frequency, the bandpass filter width, and the bandstop filter width.

[0130] Step 260: Based on the center frequency and the width of the bandpass filter, construct a bandpass filter using a preset filter; and based on the center frequency and the width of the bandstop filter, construct a bandstop filter using a preset filter.

[0131] Steps 250-260 are described uniformly as follows:

[0132] In this implementation, for each audio block, initialization parameters, such as the notch frequency range, are used as input parameters to calculate the core frequency parameters through bandpass filtering. The filter is instantiated and configured based on the core frequency parameter interface of the bandpass filter. The filters used in this embodiment include bandpass filters and bandstop filters.

[0133] In practice, the instantiation configuration of the filter is based on the reference frequency. When instantiating the filter, the core frequency parameters used by different filter instantiations can be different. The instantiation configuration is based on the preset level of filter and is configured through the core frequency parameters to form the corresponding bandpass filter and bandstop filter.

[0134] Specifically, the instantiation of bandpass / bandstop filters can be implemented using corresponding filters. In this embodiment, the preset filters may include, but are not limited to, Butterworth filters, Chebyshev filters, etc. Taking the instantiation configuration of a Butterworth filter as an example, a 5th-order Butterworth filter can be used, with the center frequency and bandpass filter width as configuration parameters. It is configured to pass frequencies within the range of centerFreq (center frequency) and width (bandpass filter width). The constructed bandpass filter is used to process subsequently added Gaussian noise, so that it only contains components of the notch edge frequency band.

[0135] A 5th-order Butterworth filter is used, with the center frequency and band-stop filter width as configuration parameters, configured to attenuate frequencies within the range of centerFreq and stopWidth (band-stop filter width). The band-stop filter constructed in this way can achieve a core-cutting effect.

[0136] In one optional embodiment, taking initialization parameters and an audio block as input, the core frequency parameters are determined through bandpass filtering. Specifically, this may include taking the notch audio range in the initialization parameters as input, and based on… Reference frequency for calculating the gain range of the notch edge ;according to Calculate the center frequency of the center position of the notch band. ;according to The coverage of the notch frequency band and edge gain region is determined to obtain the bandpass filter width. ;according to Calculate the width of the band-stop filter It is used to accurately locate the core frequency band of the cut that needs to be attenuated.

[0137] Step 270: For the audio block, key target noise is introduced using random noise and a bandpass filter, and the target noise is superimposed on the audio block signal using optimal gain to obtain gain information.

[0138] In one optional embodiment, this embodiment introduces key target noise using random noise and a bandpass filter for an audio block, and then uses optimal gain to superimpose the target noise with the audio block signal to obtain gain information. This may include: generating random noise for the audio block and processing the random noise through a bandpass filter to obtain key target noise, the target noise containing frequency components that retain only the frequency bands at the cut edges; determining the optimal gain based on the edge gain in the initialization parameters and the debugging information of the target user in the visualized spectrum; and using the optimal gain to superimpose the target noise with the original signal corresponding to the audio block to obtain gain information, the gain information being used to provide energy compensation for the cut edges.

[0139] In practical implementation, a Gaussian distribution algorithm can be used to generate random noise that conforms to a Gaussian distribution. Then, a bandpass filter is used to process the random noise so that only the frequency components of the notch edge band are retained, avoiding the introduction of interference from irrelevant frequency bands, thus obtaining the key target noise.

[0140] Then, an optimal gain is introduced into the superposition of the target noise and the original signal. Considering that the gain of the noise component is controlled by the edge gain set by the subject, the optimal gain is the gain parameter preset by the subject. When the subject sets the edge gain parameter, a visual comparison of the original spectrum and the spectrum after notching (bandpass filtering) of the example audio can be provided, allowing the user to listen to the final effect and accurately lock in the optimal gain.

[0141] In this embodiment, during the processing of each audio block, the filtered target noise will be superimposed on the original signal (i.e., the signal corresponding to the audio block) to provide energy compensation for the cutting edge.

[0142] Therefore, this embodiment improves the generation process of notched music. First, it introduces streaming processing to achieve a streaming experience where processing and playback occur simultaneously. For pure tones generated by tinnitus testing, this embodiment applies short envelopes (approximately 2ms) at the beginning and end of the waveform during generation, effectively avoiding click sounds caused by discontinuous amplitude at the beginning and end during loop playback. Then, a filter is introduced to add crucial Gaussian noise containing only the notched edge frequency bands, while simultaneously achieving the core notched effect. Finally, optimal gain is introduced to provide energy compensation for the notched edges.

[0143] Step 280: Based on the cutting mode and channel mode selected by the target user, real-time processing and cutting enhancement are performed according to the band-stop filter and gain information to generate an audio signal that meets the requirements of acoustic intervention, which is used as cutting music.

[0144] In one optional embodiment, this embodiment performs real-time processing and chip enhancement based on the chipping mode and channel mode selected by the target user, according to the band-stop filter and gain information, to generate an audio signal that meets the acoustic intervention requirements, as chipped music. Specifically, it may include: performing real-time processing and chip enhancement based on the chipping mode selected by the target user, according to the band-stop filter and gain information, to generate chipped music that meets the acoustic intervention requirements; and outputting chipped music in the corresponding channel based on the channel mode selected by the target user; wherein, during the chipped music output process, after the chipped music of the current audio block is played, the next audio block is processed immediately, until the chipped music of all audio blocks is played.

[0145] In its implementation, this embodiment achieves personalized notching processing in multiple modes and provides corresponding listening functions in the gain setting. The core principle of notching processing in different modes is similar. The following is an introduction using binaural notching as an example:

[0146] In binaural notching mode, the following operations are performed on the left and right channels respectively: ① The noise after bandpass filtering is superimposed on the original audio data with edgeGain gain to enhance the frequency band energy at the edge of the notch; ② The audio signal of the core notch frequency band (from centerFreq to stopWidth) is attenuated by a bandstop filter to achieve the "notch" effect.

[0147] For example, the processing logic for real-time processing and notch enhancement based on band-stop filter and gain information is as follows:

[0148] Left channel output = Band-stop filter (original left channel + edge gain × bandpass noise)

[0149] Right channel output = Band-stop filter (original right channel + edge gain × bandpass noise)

[0150] It should be noted that the tinnitus intervention scheme in this embodiment does not limit the specific notch pattern and duct pattern. For example, the notch pattern can include at least the following modes:

[0151] ① Mono: After notching, only the channel on the side with tinnitus is retained, while the other side is set to zero, thus achieving directional processing;

[0152] ② Single ear notching: Notching is performed only on the tinnitus-side audio channel, while the non-tinnitus-side audio remains unchanged.

[0153] ③ Bilateral notching: Notching is performed on both sides of the tinnitus duct;

[0154] ④ Local frequency band notches;

[0155] ⑤ No cutting: No processing is done, and the original audio is played in both ears.

[0156] In this embodiment, the audio data that has completed the entire process of cutting music is written to the local cache, and a double buffering mechanism is adopted (audio block chunk_a.wav and audio block chunk_b.wav are used alternately) to avoid file conflicts during the player reading process.

[0157] During the notch-based music output, the playback progress stream can be monitored, and the next segment is started immediately after each chunk finishes playing. Loop playback and interrupted resumption are supported. The final output is a notch-enhanced dual-channel audio signal that has undergone real-time processing and can be played synchronously in therapeutic settings.

[0158] As can be seen, the embodiments of this application improve the processing flow of notched music. This processing flow achieves customized processing of notched music through a combination of "band-stop filter attenuation of the core frequency band + band-pass noise enhancement of the edge frequency band": the band-stop filter precisely cuts off the frequency band where tinnitus is located, while the energy of the notched edge is compensated by band-pass noise and edge gain, forming an audio signal that meets the needs of acoustic intervention and is adapted to the personalized treatment scenarios of tinnitus patients. Unlike traditional notched music software or devices, the edge gain and notched frequency of the notched music processing algorithm used in this embodiment are adjustable in real time. Users can adjust the edge gain and notched frequency in real time and listen to different settings to find the most suitable parameters for themselves.

[0159] Step 290: Monitor the target user's neural activity state in real time through a brain-computer interface, combine notch music with neural feedback information, and adjust the notch music based on neural feedback information. At the end of the tinnitus intervention, obtain the training result of the current tinnitus intervention training.

[0160] In its specific implementation, this embodiment monitors the target user's neural activity state in real time via a brain-computer interface, combines notched music with neural feedback information, and adjusts the notched music based on the neural feedback information. At the end of the tinnitus intervention, the training result of the current tinnitus intervention training is obtained. Specifically, this may include: replacing the background music used in the neural feedback scenario with notched music; monitoring the target user's neural activity state in real time via a brain-computer interface to obtain neural feedback information; adjusting the final volume of the notched music output based on the neural feedback information; determining the practice data for the current cycle of tinnitus intervention based on the target user's tinnitus loudness measurement information at the end of the tinnitus intervention; and determining the training result of the tinnitus intervention training based on the practice data of the current cycle and the practice data of the previous cycle.

[0161] In practical implementation, notched music can be combined with neurofeedback scenarios. Neurofeedback information is obtained by monitoring the subject's neural activity state (such as meditation, concentration, etc.) through a brain-computer interface.

[0162] Taking a meditation scenario as an example, during meditation, the background music in the meditation practice scene is replaced with notch music, and the volume of the notch music is adjusted according to the meditation score (mainly determined based on neurofeedback information). The higher the meditation score, the louder the background volume, the better it masks tinnitus, and the more relaxed and comfortable the user feels.

[0163] The final volume during music playback is calculated as: system volume + adjustment coefficient * meditation score. Preferably, the meditation score ranges from 0 to 100, and the adjustment coefficient ranges from 0.1 to 1, allowing for user-adjusted settings. The final volume can be limited to 0-100.

[0164] Furthermore, in practice, subjects can adjust notch music parameters in real time, including tinnitus frequency and edge gain. For users with different tinnitus levels in their left and right ears, the notch parameters for the left and right channels can be adjusted separately in real time. The volume for each ear can also be adjusted independently. At the beginning and end of the scenario, users can measure tinnitus loudness as practice data for the current tinnitus intervention cycle. Combined with the practice data from the previous cycle, the training results of the tinnitus intervention training are determined. For example, by comparing the loudness measured before and after scenario practice, a practice effectiveness score can be evaluated as the training result. Practice effectiveness score = Tinnitus loudness before practice - Tinnitus loudness after practice.

[0165] In summary, this embodiment first uses pure-tone audiometry to test the target user's audiogram and hearing type, then uses a matching method to determine tinnitus input parameters such as the tinnitus center frequency and tinnitus loudness. Within a buffer, the original audio is processed in real-time or near real-time, such as through dynamic streaming, to generate audio blocks. These blocks are then combined with the tinnitus input parameters for bandpass filtering to construct bandpass / bandstop filters. Next, based on the tinnitus input parameters, initialization parameters are constructed by combining multiple user-preset parameters and algorithm parameters. During preprocessing, the original audio is dynamically processed using streaming to form a two-dimensional matrix of audio blocks. Based on the initialization parameters and audio blocks, the core parameters of the bandpass filter are calculated, thereby instantiating and configuring the bandpass / bandstop filter. Subsequently, for notching processing, algorithms such as Gaussian distribution are introduced to generate random noise. This random noise is processed by the bandpass filter, and key target noise is analyzed. User-preset parameters are then locked as the optimal gain, which is applied to the superposition of the target noise and the audio block signal to obtain gain information that provides energy compensation. Based on the notch pattern and tract pattern selected by the target user, real-time processing and notch enhancement are performed using band-stop filters and gain information to generate notch music corresponding to audio blocks that meet the acoustic intervention requirements. Finally, the notch music is combined with a neurofeedback scenario, and the target user's neural activity is monitored in real-time via a brain-computer interface. This allows the notch music to be adjusted based on the neurofeedback information, and the training result of the current tinnitus intervention is obtained at the end of the intervention. Therefore, this application proposes an improved scheme for digital tinnitus intervention, mainly combining brain-computer interfaces, cognitive behavioral therapy, and notch music therapy to achieve complementary advantages. This makes the tinnitus intervention scheme more closely aligned with individual neurofeedback characteristics and tinnitus acoustic needs, overcoming the limitations of fixed parameters in traditional schemes and improving the personalized adaptability of the intervention scheme. Furthermore, through technological integration and digital innovation, while improving user operation convenience and data tracking capabilities, it breaks through the scenario limitations of traditional schemes, providing a more flexible and accessible solution for the field of tinnitus intervention, thereby solving problems such as untimely audio processing and poor tinnitus intervention effects in existing digital tinnitus intervention technologies.

[0166] Furthermore, this embodiment also achieves: ① Flexible mode adaptation. Unlike the notching processing method in existing technologies, this embodiment adds channel processing on the basis of notching processing. Users can compare and choose processing modes such as mono, mono-ear notching, binaural notching, and no notching according to their own situation, and combine them with neurofeedback meditation. Mono and mono-ear notching can focus the treatment on the ear with tinnitus for patients with unilateral severe tinnitus, avoiding the healthy ear from being affected. The no-notching mode can be used to allow users to carefully compare the results before and after processing, assisting in the precise selection of edge gain parameters and generating the best notching scheme. ② Tinnitus quantification and tracking. This embodiment can measure meditation score, audiogram (through pure tone audiometry), tinnitus frequency, tinnitus loudness, and edge gain. Through continuous use over multiple days, the dynamic changes of these indicators can be measured, increasing the user's sense of control and compliance.

[0167] Furthermore, the solution in this embodiment also includes:

[0168] Comprehensive assessment and clinical application value:

[0169] ① Dynamic effect assessment capability: By comparing the tinnitus loudness before and after meditation, the effect of a single intervention can be reflected in real time, providing an intuitive reference for adjusting personalized plans and helping to optimize and iterate intervention strategies;

[0170] ② Standardized data collection mode: Integrating multi-dimensional quantitative indicators to form a systematic tracking system, which not only facilitates users' self-monitoring, but also provides a standardized data collection model for clinical research.

[0171] Cost optimization and accessibility expansion:

[0172] Lightweight hardware requirements: No professional medical equipment is required. The entire process of hearing testing and intervention can be completed using only a mobile phone and ordinary headphones, effectively reducing hardware costs and the barrier to entry. It is especially suitable for users in areas with limited resources or those with mobility impairments.

[0173] It should be noted that, for the sake of simplicity, the method embodiments are described as a series of actions. However, those skilled in the art should know that the embodiments of this application are not limited to the described order of actions, because according to the embodiments of this application, some steps may be performed in other orders or simultaneously.

[0174] like Figure 3 As shown in the figure, this application embodiment also provides a tinnitus intervention device 300 based on a brain-computer interface, including:

[0175] Pure tone audiometry module 310 is used to perform the steps of generating audiograms and audio types;

[0176] Frequency loudness measurement module 320 is used to perform the steps of measuring tinnitus frequency and tinnitus loudness;

[0177] Preprocessing module 330 is used to perform real-time or near-real-time processing of the raw audio data;

[0178] The bandpass filter processing module 340 is used to perform the bandpass filter processing on the audio block;

[0179] The noise processing module 350 is used to perform the step of superimposing the target noise and the audio block with the optimal gain;

[0180] The track-cut music generation and output module 360 ​​is used to perform track-cutting processing and output based on the track-cutting mode and channel mode selected by the target user;

[0181] The tinnitus intervention training result determination module 370 is used to perform the steps of real-time monitoring of the target user's neural activity state and adjusting the notch music based on neural feedback information.

[0182] It should be noted that the tinnitus intervention device based on brain-computer interface provided in the embodiments of this application can execute the tinnitus intervention method based on brain-computer interface provided in any embodiment of this application, and has the corresponding functions and beneficial effects of the execution method.

[0183] In practical implementation, the aforementioned brain-computer interface-based tinnitus intervention device can be integrated into a device, such as an earphone device. This allows the device to dynamically generate notched music in real time based on the user's pure tone test results, combining it with brain-computer interface meditation for tinnitus intervention. As an electronic device, it integrates brain-computer interface, cognitive behavioral therapy, and notched music therapy to form a digital, interactive comprehensive tinnitus intervention program. This electronic device can consist of two or more physical entities, or it can consist of a single physical entity. For example, the electronic device can be a personal computer (PC), a computer, a server, etc. This application embodiment does not impose specific limitations in this regard.

[0184] In a specific implementation, this application provides an electronic device with data processing, storage, and communication functions, used to implement the steps of the brain-computer interface-based tinnitus intervention method provided in any of the above embodiments. For example, refer to... Figure 4The electronic device may include a processor 111, a communication interface 112, a memory 113, and a communication bus 114. The processor 111, communication interface 112, and memory 113 communicate with each other via the communication bus 114. The memory 113 stores computer programs. The processor 111, when executing the program stored in the memory 113, implements the steps of the brain-computer interface-based tinnitus intervention method provided in any of the aforementioned method embodiments. For example, the method may include the following steps: obtaining an audiogram and hearing type generated by digital pure-tone audiometry of a target user, where the target user includes a group with auditory perception abnormalities requiring intervention; based on the audiogram and hearing type, performing tinnitus frequency and loudness measurements on the target user using a matching method to obtain tinnitus input parameters, including the tinnitus center frequency obtained from the tinnitus frequency measurement and the relative tinnitus loudness obtained from the tinnitus loudness measurement; within an established music buffer, performing real-time or near-real-time processing based on the original audio data, and combining the initialization parameters constructed based on the tinnitus input parameters with pre-processed audio blocks represented by a two-dimensional matrix for bandpass processing. The process involves filtering, constructing bandpass and bandstop filters. For each audio block, random noise and a bandpass filter are used to introduce key target noise, which is then superimposed with the audio block using optimal gain to obtain gain information. Based on the notch pattern and tract pattern selected by the target user, real-time processing and notch enhancement are performed according to the bandstop filter and gain information to generate an audio signal that meets the acoustic intervention requirements, serving as notch music. The neural activity state of the target user is monitored in real time through a brain-computer interface, and the notch music is combined with neural feedback information to adjust the notch music based on the neural feedback information. At the end of the tinnitus intervention, the training results of the current tinnitus intervention training are obtained.

[0185] This application also provides a computer-readable storage medium storing a computer program thereon, which, when executed by a processor, implements the steps of the tinnitus intervention method based on a brain-computer interface as provided in any of the foregoing method embodiments.

[0186] It should be noted that, in this document, relational terms such as "first" and "second" are used merely to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitations, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes said element.

[0187] The above description is merely a specific embodiment of this application, enabling those skilled in the art to understand or implement this application. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of this application. Therefore, this application is not to be limited to the embodiments shown herein, but is to be accorded the widest scope consistent with the principles and novel features claimed herein.

Claims

1. A tinnitus intervention device based on a brain-computer interface, used to perform the following steps, characterized in that, include: Obtain audiograms and hearing types generated from digital pure-tone audiometry of target users, including groups with auditory perception abnormalities requiring intervention; Based on audiograms and hearing types, tinnitus frequency and tinnitus loudness are measured for target users using a matching method to obtain tinnitus input parameters. The tinnitus input parameters include the tinnitus center frequency obtained from the tinnitus frequency measurement and the relative tinnitus loudness obtained from the tinnitus loudness measurement. Within the established music buffer, real-time or near-real-time processing is performed on the raw audio data. Based on the initialization parameters constructed from the tinnitus input parameters, and combined with the pre-processed audio blocks represented by a two-dimensional matrix, bandpass filtering is performed to construct bandpass and bandstop filters. For the audio block, random noise and a bandpass filter are used to introduce key target noise, and the target noise is superimposed with the audio block signal using the optimal gain to obtain gain information; Based on the cutting mode and channel mode selected by the target user, real-time processing and cutting enhancement are performed according to the band-stop filter and gain information to generate an audio signal that meets the requirements of acoustic intervention, which is then used as cutting music. By monitoring the target user's neural activity state in real time through a brain-computer interface, the notch music is combined with neural feedback information, so that the notch music is adjusted based on the neural feedback information. At the end of the tinnitus intervention, the training result of the current tinnitus intervention training is obtained. The process involves real-time monitoring of the target user's neural activity via a brain-computer interface, combining notched music with neurofeedback information, and adjusting the notched music based on this information. At the end of the tinnitus intervention, the training result is obtained, including: replacing the background music used in the neurofeedback scenario with notched music; real-time monitoring of the target user's neural activity via the brain-computer interface to obtain neurofeedback information; adjusting the final volume of the notched music output based on the neurofeedback information; determining the practice data for the current tinnitus intervention cycle based on the target user's tinnitus loudness measurement information at the end of the tinnitus intervention; and determining the training result of the tinnitus intervention training based on the current cycle's practice data and the previous cycle's practice data. The neurofeedback scenarios include meditation scenarios and attention training.

2. The apparatus according to claim 1, characterized in that, Based on audiograms and hearing types, a matching method is used to measure the tinnitus frequency and loudness of the target user, obtaining tinnitus input parameters, including: Based on the frequency range, the frequency coordinate system is divided into average laws to construct a frequency adjustment benchmark; When the target user conducts a pure tone test, based on the frequency adjustment benchmark, the pitch adjustment information of the target user through interactive operation is determined, and the frequency of the pitch adjustment information is taken as the tinnitus frequency corresponding to the user's perception of tinnitus sound. The matching method is used to compare the tinnitus frequency with pure tones of known frequencies to determine the tinnitus center frequency; Identify the volume adjustment information of the target user through interactive operations, and determine the tinnitus volume to mask the tinnitus sound based on the volume adjustment information; The relative tinnitus loudness is determined by comparing the tinnitus volume with a volume baseline percentage using a matching method.

3. The apparatus according to claim 1, characterized in that, Within the established music buffer, real-time or near-real-time processing is performed on the raw audio data. Initialization parameters, constructed based on tinnitus input parameters, are combined with pre-processed audio blocks represented by two-dimensional matrices to perform bandpass filtering, constructing bandpass and bandstop filters, including: The target user's preset tinnitus side parameters and processing mode, combined with the tinnitus input parameters, are used as user setting parameters. The obtained algorithm initialization parameters and user setting parameters are used as the initialization parameters for the notch music generation algorithm. According to the preset time segments, the raw audio data currently being played is preprocessed in real time or near real time to obtain audio blocks represented by a two-dimensional matrix; Using initialization parameters and audio blocks as input, the core frequency parameters are determined through filtering and bandpass processing. The core frequency parameters include at least the reference frequency of the notch, the center frequency, the bandpass filter width, and the bandstop filter width. Based on the center frequency and the width of the bandpass filter, a bandpass filter is constructed using a preset filter; and based on the center frequency and the width of the bandstop filter, a bandstop filter is constructed using a preset filter. Before processing the raw audio data, a music buffer is constructed, and the raw audio data is loaded into a memory matrix and cached. During the processing of the raw audio data, each piece of data processed is an audio block in the raw audio data, and each audio block is determined by the segment index and segment size.

4. The apparatus according to claim 3, characterized in that, Using initialization parameters and audio blocks as input, the core frequency parameters are determined through bandpass filtering, including: Taking the notch audio range in the initialization parameters as input, according to Reference frequency for calculating the gain range of the notch edge ; according to Calculate the center frequency of the center position of the notch band. ; according to The coverage of the notch frequency band and edge gain region is determined to obtain the bandpass filter width. ; according to Calculate the width of the bandstop filter It is used to accurately locate the core frequency band of the cut that needs to be attenuated.

5. The apparatus according to claim 1, characterized in that, For the audio block, random noise and a bandpass filter are used to introduce key target noise. The target noise is then superimposed with the audio block signal using optimal gain to obtain gain information, including: For an audio block, random noise is generated and processed by a bandpass filter to obtain the key target noise, which contains frequency components that retain only the frequency bands at the edges of the notch. Based on the edge gain in the initialization parameters, the optimal gain is determined according to the debugging information of the target user in the visualized spectrum; The target noise is superimposed with the original signal corresponding to the audio block using the optimal gain to obtain gain information, which is used to provide energy compensation for the cut edge.

6. The apparatus according to claim 1, characterized in that, Based on the chipping mode and channel mode selected by the target user, real-time processing and chipping enhancement are performed according to band-stop filters and gain information to generate audio signals that meet the requirements of acoustic intervention, as chipped music, including: Based on the cleaving pattern selected by the target user, real-time processing and cleaving enhancement are performed according to the band-stop filter and gain information to generate cleaving music that meets the requirements of acoustic intervention. Based on the channel mode selected by the target user, output cut music in the corresponding channel; During the output of the cut music, after the cut music of the current audio block finishes playing, the next audio block is processed immediately, until the cut music of all audio blocks has finished playing.

7. The apparatus according to claim 1, characterized in that, Obtain the audiogram and hearing type generated from the digital pure-tone audiometry of the target user, including: During the digital pure-tone audiometry process for target users, the pure tone is output from low to high frequencies through air conduction audiometry. According to the threshold determination algorithm, the air conduction threshold is determined based on the feedback operation of the target user during pure tone output. The air conduction threshold is marked on the preset pure tone audiogram to generate an audiogram. Hearing loss is then analyzed based on the air conduction threshold to determine the hearing type.

8. A tinnitus intervention device based on a brain-computer interface, characterized in that, include: The pure-tone audiometry module is used to perform the steps of generating audiograms and audio types. The frequency and loudness measurement module is used to perform the steps of measuring tinnitus frequency and tinnitus loudness. The preprocessing module is used to perform real-time or near-real-time processing on the raw audio data; The bandpass filtering module is used to perform the bandpass filtering process on the audio block; The noise processing module is used to perform the step of superimposing the target noise and the audio block into signals using optimal gain. The track-cutting music generation and output module is used to perform track-cutting processing and output based on the track-cutting mode and channel mode selected by the target user; The tinnitus intervention training result determination module is used to perform the steps of real-time monitoring of the target user's neural activity state and adjusting the notch music based on neural feedback information; The process involves real-time monitoring of the target user's neural activity via a brain-computer interface, combining notched music with neurofeedback information, and adjusting the notched music based on this information. At the end of the tinnitus intervention, the training result is obtained, including: replacing the background music used in the neurofeedback scenario with notched music; real-time monitoring of the target user's neural activity via the brain-computer interface to obtain neurofeedback information; adjusting the final volume of the notched music output based on the neurofeedback information; determining the practice data for the current tinnitus intervention cycle based on the target user's tinnitus loudness measurement information at the end of the tinnitus intervention; and determining the training result of the tinnitus intervention training based on the current cycle's practice data and the previous cycle's practice data. The neurofeedback scenarios include meditation scenarios and attention training.

Images