A method and system for binaural hearing aid dual channel enhanced audio intelligibility directivity

Abstract

The present application relates to the technical field of hearing aids, in particular to a method and system for enhancing the audio clarity directivity of a hearing aid in two channels, comprising: sending a detection signal to identify a mobile phone power supply protocol, receiving and identifying a UAC protocol, outputting corresponding safe voltage and current according to the type of mobile phone power supply protocol, triggering three-level protection, identifying the current audio output mode, sampling rate and bit depth of the mobile phone by analyzing the UAC protocol, and obtaining a standard audio stream format; receiving a scene mode instruction, controlling the dual-microphone channel according to the scene mode instruction to collect analog signals, and converting the analog signals into digital audio signals with source identification respectively; generating enhanced digital audio according to the type of digital audio signal; reading the connection state parameters fed back by the mobile phone in real time, and encapsulating the enhanced digital audio into a Bluetooth signal according to the connection state parameters and transmitting the Bluetooth signal to a Bluetooth hearing aid. The present application realizes high-precision audio enhancement by combining dynamic adaptive filtering with multi-frequency band regulation.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of hearing aids, in particular to a method and system for enhancing the audio clarity directivity of a hearing aid in two channels. BACKGROUND

[0002] In the application scenarios of hearing aid audio processing and wireless transmission, the hearing aid often needs to be connected with Android phones of different brands and models through a TypeC interface to realize power supply, audio data interaction and scene control cooperation. However, there is a significant heterogeneity problem between the phone and the hearing aid, mainly manifested in the differences in power supply protocols (supporting PD protocol or QC protocol), audio output mode heterogeneity (digital signal "D" or analog signal "A") and connection state fluctuation (stuttering frequency, transmission rate difference up to 2 times or more). Since the power supply protocol determines the power supply voltage and current output standard, the audio mode and the connection state directly affect the signal processing adaptability, and the power supply matching degree, signal compatibility and transmission stability of heterogeneous devices under a unified interaction process differ greatly, resulting in high difficulty in parameter cooperation in the whole process.

[0003] Traditional hearing aid audio processing methods mostly use fixed parameter processes and do not consider the differences in the heterogeneity characteristics of the phone, which can easily cause some phones to have insufficient power supply or overcurrent in the core modules (phone audio conversion module, microphone audio conversion module) due to mismatched power supply protocols, signal distortion in some phones due to audio mode heterogeneity, and Bluetooth transmission stuttering and audio-visual asynchrony due to connection state fluctuation. At the same time, the working environment of the hearing aid is complex, and factors such as environmental electromagnetic interference (such as 2.4GHz frequency band Wi-Fi signal interference), temperature and humidity fluctuations (such as outdoor high temperature and humidity environment), and device vibration (such as collision during user movement) can introduce current monitoring noise, audio sampling deviation and Bluetooth transmission packet loss, further reducing power supply safety and audio processing accuracy.

[0004] In addition, the existing system lacks dynamic adaptation capability to the heterogeneity characteristics of the phone, cannot quantitatively adjust the power supply parameters in real time to cope with current fluctuations (such as sudden overcurrent), and is also difficult to dynamically optimize the microphone sound collection strategy and audio enhancement algorithm for different scenes (meeting / outdoor); and temperature control adjustment mostly relies on fixed filtering and transmission protocols, with a lag response to signal deviation and connection state changes, making it difficult to achieve high-precision requirements such as enhanced audio signal-to-noise ratio ≥55dB and Bluetooth transmission delay ≤100ms within 100ms, which seriously affects the audio clarity of the hearing aid and the user experience, and restricts the compatibility of the production line to diversified phone devices and the expansion of product application scenarios. SUMMARY

[0005] The present application realizes high-precision audio enhancement through dynamic adaptive filtering and multi-frequency band regulation.

[0006] The technical scheme provided by the application is as follows: a method for enhancing the audio clarity directivity of a hearing aid in two channels, the method comprising:

[0007] The power processing module sends a detection signal to identify the mobile phone power supply protocol, the mobile phone audio conversion module receives and identifies the UAC protocol, outputs corresponding safe voltage and current according to the type of the mobile phone power supply protocol, judges whether the power supply power meets the demand of the core module, monitors the current in real time, triggers three-level protection according to the current deviation, identifies the current audio output mode, sampling rate and bit depth of the mobile phone by analyzing the UAC protocol, and obtains the standard audio stream format by using a general adaptive algorithm for normalization;

[0008] The scene mode instruction is received, the dual-microphone channel control is performed according to the scene mode instruction to collect analog signals, and the collected analog signals are respectively converted into digital radio signals with source identification through analog-to-digital conversion;

[0009] The type of the digital radio signal is judged, if it is a double-identification signal, the time domain delay of the two signals is calculated, the frequency domain characteristic difference is analyzed to distinguish language and environmental noise, a filtering model is constructed by taking the device microphone signal as the core and combining the noise characteristics of the mobile phone microphone signal, the interference signal is removed to obtain enhanced digital audio, and if it is a single-identification signal, the enhanced digital audio is generated through a wind and noise prevention algorithm;

[0010] The connection state parameters fed back by the mobile phone are read in real time, and the enhanced digital audio is packaged into Bluetooth signals in different ways according to the connection state parameters and transmitted to the Bluetooth hearing aid.

[0011] Preferably, the specific process of judging whether the power supply power meets the demand of the core module is as follows:

[0012] The TypeC interface is physically connected with the mobile phone, the power processing module sends a detection signal to identify the mobile phone power supply protocol through the CC line, outputs corresponding standard power supply voltage and safe power supply current according to the identified mobile phone power supply protocol, if it is a PD protocol, outputs 5V voltage and 2A current, if it is a QC protocol, outputs 9V voltage and 2A current; the power processing module multiplies the output power supply voltage and current according to the power supply power calculation formula to obtain the actual power supply power; the calculated actual power supply power is compared with the total power consumption of the device core module to judge whether the current power supply power can meet the operation demand of the core module.

[0013] Preferably, the three-level protection has the following specific contents:

[0014] The power processing module monitors the power supply current in real time, calculates the current deviation , and triggers three-level protection according to the current deviation and duration; when When the current exceeds the threshold, the first level protection is triggered, the current fluctuation warning signal is sent to the mobile phone through the TypeC interface, and the warning log is buffered in the device. , and the duration exceeds 50 , the second level protection is triggered, the protection control unit sends instructions to the power management chip, the output voltage is fine-tuned through PWM technology, the current is reversely controlled to return to the stable range, and the adjustment parameters are updated every 10 times. If the adjustment is invalid for three times, the third level protection is triggered. , and the duration exceeds 50 , the third level protection is triggered, the power supply circuit is immediately cut off, the fault is locked, and the current, voltage and duration at the time of overcurrent are stored in the non-volatile memory. Before power off, a pre-warning signal is sent to the core module to save intermediate data.

[0015] Preferably, the specific obtaining process of the standard audio stream format is as follows:

[0016] If it is identified as a digital PCM signal, the general PCM decoding module performs standardization processing. If the original sampling rate is 44100 Hz and the original bit depth is 16 bits, the sampling rate is increased to 48000 Hz through linear interpolation algorithm, and the bit depth is expanded to 24 bits through high-bit zero padding. If the original parameters are 48000 Hz and 24 bits, the parameters are directly retained to generate a standard digital audio stream format. If it is identified as an analog AC coupled signal, the original analog signal is sampled through a 16-bit resolution ADC to obtain the original amplitude, an automatic gain control module is started, the original amplitude is normalized to the interval of -1V +1V, and the normalized analog signal is converted into a digital signal through the ADC. The sampling rate is set to 44100 Hz and the bit depth is set to 16 bits to generate a standard analog-to-digital audio stream format.

[0017] Preferably, the specific process of obtaining the digital radio signal is as follows:

[0018] The scene mode command that has passed CRC verification is extracted from the temporary parameter buffer. The scene mode command type is determined. If it is a conference mode, the mobile phone microphone and the device's built-in microphone are activated simultaneously. The mobile phone microphone collects the original analog signal of ambient sound at a sampling rate of 44.1kHz and a bit depth of 16 bits, while the device's built-in microphone collects the original analog signal of target voice at a sampling rate of 48kHz and a bit depth of 24 bits. If it is an outdoor mode, a shutdown signal is sent to the mobile phone microphone, and only the device's built-in microphone is activated. This microphone collects at a sampling rate of 48kHz and a bit depth of 24 bits, and pre-filters low-frequency wind noise below 500Hz using an adaptive filtering algorithm to generate a wind noise-reduced original analog signal. The microphone audio conversion module starts analog-to-digital conversion. The ADC parameters in the module are precisely matched with the corresponding microphone acquisition parameters. For the original analog signal of ambient sound collected by the mobile phone microphone, the ADC is converted at a sampling rate of 44.1kHz and a resolution of 16 bits to generate a digital audio signal with the mobile phone microphone source identifier. For the original analog signal of target voice and the wind noise-reduced original analog signal collected by the device's built-in microphone, the ADC is converted at a sampling rate of 48kHz and a resolution of 24 bits to generate digital audio signals with the device microphone source identifier.

[0019] Preferably, the enhanced digital audio acquisition process is as follows:

[0020] If the input is a digital audio signal with dual identifiers for both the mobile phone microphone and the device microphone, the sampling rate of the mobile phone microphone signal is increased to 48kHz through an interpolation algorithm. The time-domain delay of the two signals is calculated to locate the direction of the target sound. The frequency domain feature difference in the 1kHz-3kHz band is analyzed to distinguish speech from environmental noise. Taking the device microphone signal as the core, an adaptive filtering model is constructed by combining the noise feature of the mobile phone microphone signal with a frequency domain feature difference of less than 5%. The interference signal in the device microphone signal that matches the noise template is dynamically attenuated. After amplitude normalization and 24-bit resampling, enhanced digital audio is generated. If the input is a digital audio signal with only a single identifier for the device microphone, the wind noise reduction algorithm is directly activated. The Butterworth high-pass filter attenuates the low-frequency wind noise band below 500Hz by 10-15dB and dynamically gains the core speech band of 300Hz-3kHz by 3-5dB to generate enhanced digital audio.

[0021] Preferably, the Bluetooth signal encapsulation process is as follows:

[0022] The connection status parameters fed back by the mobile phone are read in real time through the Type-C interface, and the frequency of lag is detected. times / minute or transmission rate At Mbps, the Bluetooth transmission protocol is switched from APT-X to SBC. If the input is enhanced digital audio in conference mode, the enhanced digital audio is encapsulated into an SBC format Bluetooth signal according to the SBC encoding format. If the input is enhanced digital audio in outdoor mode, it is encapsulated into an SBC format Bluetooth signal according to the same SBC encoding format. When the stuttering frequency... times / minute and transmission rate At Mbps, the APT-X protocol is maintained. If the input is enhanced digital audio in conference mode, the enhanced digital audio will be encapsulated into an APT-X format Bluetooth signal according to the APT-X encoding format. If the input is enhanced digital audio in outdoor mode, the enhanced digital audio will be encapsulated into an APT-X format Bluetooth signal according to the same APT-X encoding format.

[0023] The present invention also provides a system for enhancing the directivity of dual-channel audio clarity in a hearing aid, the system being used to perform the aforementioned method for enhancing the directivity of dual-channel audio clarity in a hearing aid.

[0024] The present invention also provides a computer-readable storage medium storing a computer program that is executed by a processor to implement the method for enhancing the directivity of dual-channel audio clarity in a hearing aid.

[0025] The beneficial effects of this invention are:

[0026] 1. The system receives "Meeting Mode (M1) / Outdoor Mode (M2)" scenario commands via the Type-C interface to achieve differentiated control of the dual microphone channels. In M1 mode, the mobile phone microphone and the device's built-in microphone are activated simultaneously to collect ambient sound and target speech at 44.1kHz / 16bit and 48kHz / 24bit respectively, and the collected parameters are precisely matched with the standard parameters of the mobile phone's audio conversion module. In M2 mode, the mobile phone microphone with weak wind noise resistance is turned off, and only the device's built-in microphone is activated. Low-frequency wind noise below 500Hz is pre-filtered through adaptive filtering. At the same time, the mobile phone microphone signal is used as a noise reference source, and the noise sub-band is marked by the energy ratio difference in the 1kHz-3kHz frequency band. Combined with power spectrum similarity, the system makes dual judgments on interference, which not only achieves accurate adaptation to different scenarios, but also significantly reduces the probability of speech false suppression and noise omission suppression. The ambient noise attenuation in the meeting scenario can reach 5-10dB, and the clarity of the target speech is significantly improved. At the same time, it avoids signal distortion caused by parameter incompatibility, providing high-quality input for subsequent AI processing.

[0027] 2. Based on the device's microphone signal, an adaptive filtering model is constructed by combining the noise characteristics of the mobile phone microphone. Differential attenuation of interference subbands of varying intensities is achieved through dynamic suppression gain (adjustable from 5-10dB). Strong interference is deeply suppressed, while weak interference is moderately attenuated, preserving the target speech details to the greatest extent possible and enhancing the audio signal-to-noise ratio (SNR) to ≥60dB. For single signals in M2 mode, a combination algorithm of "time-domain LMS filtering" and "frequency-domain notch filtering" is used to attenuate wind noise below 500Hz by 10-15dB, while dynamically increasing the speech frequency band from 300Hz to 3kHz by 3-5dB to compensate for speech attenuation, solving the problem of "noise suppression resulting in speech loss" in traditional filtering. Furthermore, signal reconstruction is achieved through 2048-point FFT / IFFT and 50% overlap addition, preserving the original phase information. Amplitude normalization ensures a unified format, guaranteeing both the clarity and integrity of the enhanced audio and providing standardized input for Bluetooth transmission.

[0028] 3. Real-time reading of mobile phone connection status parameters (stuttering frequency, transmission rate). When the connection is stable, the APT-X protocol (576kbps) is used to ensure sound quality. When the connection is stuttering / slow, the SBC protocol (320kbps) is switched to improve packet loss resistance. Combined with FHSS frequency hopping technology, it avoids interference in the 2.4GHz band, with a transmission latency of ≤100ms to avoid audio-visual desynchronization. After decoding at the hearing aid end, the analog audio amplitude is stabilized at -1V. +1V ensures consistent sound quality. The power supply system employs a three-level protection mechanism (early warning, dynamic adjustment, and forced power-off) to monitor current fluctuations in real time. It features adaptive output via PD / QC protocol (5V / 2A, 9V / 2A), sends a warning signal 20μs before forced power-off, caches intermediate data in the core module, and can resume operation after a fault reset. Simultaneously, fault logs are stored for easy troubleshooting. This approach not only ensures power supply safety and transmission stability but also optimizes the collaborative efficiency of multiple modules, reduces initialization time, and improves device response speed and user experience. Attached Figure Description

[0029] Figure 1 This is a flowchart of a method for enhancing the directivity of audio clarity in a hearing aid using dual channels, according to the present invention.

[0030] Figure 2 This is a flowchart illustrating the processing steps of a method for enhancing the directivity and clarity of audio in a dual-channel hearing aid according to the present invention. Detailed Implementation

[0031] The following description is intended to disclose the present invention and enable those skilled in the art to implement it. The preferred embodiments described below are merely examples, and other obvious variations will occur to those skilled in the art. The basic principles of the invention defined in the following description can be applied to other embodiments, modifications, improvements, equivalents, and other technical solutions that do not depart from the spirit and scope of the invention.

[0032] It is understood that the term "a" should be understood as "at least one" or "one or more," that is, in one embodiment, the number of an element can be one, while in another embodiment, the number of the element can be multiple, and the term "a" should not be understood as a limitation on the number.

[0033] like Figure 1 and Figure 2 As shown, after the device is physically connected to the mobile phone via the Type-C interface, the power processing module immediately starts the power supply adaptation process: first, it identifies the power supply protocol type output by the mobile phone (PD protocol / QC protocol). If the PD protocol is detected, it automatically matches a safe power supply current of 10W (5V / 2A); if the QC protocol is detected, it matches a safe power supply current of 18W (9V / 2A) to ensure the stable operation of core modules such as the mobile phone audio conversion module and the microphone audio conversion module inside the device. Meanwhile, the mobile audio conversion module establishes data interaction with the mobile phone via the Type-C interface, directly parsing the USB audio universal underlying protocol (such as USBAudioClass 1.0 / 2.0, i.e., UAC protocol) transmitted through the Type-C interface. This protocol is a unified standard for audio output of all Android systems. Through protocol parsing, it automatically identifies the current audio output mode of the mobile phone (digital signal is identified by "D" / analog signal by "A"), sampling rate (44.1kHz / 48kHz), and bit depth (16bit / 24bit), and starts the universal adaptation algorithm of "underlying protocol parsing - signal normalization": if it is identified as a digital signal (identified by "D"), it calls the universal PCM decoding module to convert the signal into a standard digital audio stream with a sampling rate of 48kHz and a bit depth of 24bit; if it is identified as an analog signal (identified by "A"), it uses the automatic gain control (AGC) module to stabilize the signal amplitude at -1V. The +1V range is then converted by an ADC to a 44.1kHz sampling rate digital audio stream, ultimately generating audio conversion configuration parameters including "signal processing mode and sampling rate / bit depth parameters," providing a unified adaptation basis for subsequent audio processing. Furthermore, the bidirectional data channel of the Type-C interface synchronously receives scene mode commands ("Conference Mode" command M1 / ​​"Outdoor Mode" command M2) sent by the accompanying mobile app. These commands directly determine the working status of subsequent microphone channels (mobile phone microphone channel and device built-in microphone channel), ensuring consistency with the parameter association logic of subsequent audio control stages.

[0034] Specifically, after the device and mobile phone establish a physical connection via the Type-C interface, the power processing module first initiates power supply protocol identification: it sends a 1kHz, 3.3V protocol probe signal to the mobile phone through the CC line (Configuration Channel) of the Type-C interface, and receives the response signal from the mobile phone. If the response signal contains the "PD_ACK" identifier, it determines that the current mobile phone is using the PD (Power Delivery) power supply protocol; if it contains the "QC_ACK" identifier, it determines that it is using the QC (Quick Charge) power supply protocol. Based on the identified power supply protocol, the module matches the corresponding power supply voltage and safe current: when it is the PD protocol, it outputs a standard 5V power supply voltage (…). ) and 2A safe supply current ( When using the QC protocol, it outputs a standard 9V supply voltage. ) and 2A safe supply current ( To verify whether the power supply meets the requirements of the core modules of the equipment (mobile phone audio conversion module and microphone audio conversion module), the power supply capacity is determined using a power calculation formula, the specific formula of which is: ,in, This refers to the power supply.

[0035] During power supply, the power processing module monitors the power supply current fluctuation in real time with an accuracy of ±0.01A. If the fluctuation exceeds ±0.05A, the overcurrent protection mechanism is immediately triggered to ensure that the output voltage, current and power remain stable, providing reliable power support for the subsequent signal processing of the mobile phone audio conversion module and the sound recording of the microphone audio conversion module.

[0036] The power processing module incorporates a high-precision current sampling circuit. Through a 0.01Ω sampling resistor connected in series in the power supply circuit, it acquires the output current signal in real time (sampling frequency 10kHz), converts the analog signal into a digital signal, and transmits it to the protection control unit. The control unit operates according to the formula... Calculate current deviation ( For current deviation, For real-time current, For rated safe current, ).when When the current is stable, no action is taken; when When the alarm is triggered (first level, real-time current monitoring and abnormal alarm, no action, only alarm): on the one hand, a "current fluctuation alarm" signal is sent to the mobile phone through the TypeC interface (the signal format follows the abnormal feedback specification of the USBPD / QC protocol), and on the other hand, the alarm log is cached in the device (recording the time of fluctuation, duration, and maximum deviation value) to provide a basis for subsequent fault diagnosis, but without affecting the current power supply output, ensuring the normal operation of the mobile phone audio conversion module and the microphone audio conversion module.

[0037] If the current fluctuation continues to exceed the current fluctuation threshold ( Current fluctuation threshold (and lasting longer than 50) The protection mechanism enters the dynamic adjustment stage (second level, dynamic current regulation, active intervention, maintaining power supply). The protection control unit sends adjustment commands to the power management chip (PMIC), fine-tuning the output voltage through pulse width modulation (PWM) technology: under the PD protocol, Fine-tuning from 5V to 4.8V 5.2V range (adjustment step size 0.05V); under QC protocol, Fine-tuning from 9V to 8.8V In the 9.2V range, using the formula ( (For the current power supply) reverse control current, so that Return to A stable range of ±0.05A; during adjustment, the control unit monitors current changes in real time, every 10... Update the adjustment parameters once until the current stabilizes; if after three consecutive adjustments... If the current still exceeds 0.1A, the dynamic adjustment is deemed to have failed, triggering the third level of protection.

[0038] The protection mechanism enters the highest priority forced power-off phase (third level, forced power-off and fault locking, cutting off power supply to ensure safety) under the following two conditions: real-time current. And lasting for more than 50 (e.g., a module short circuit causing a sudden current surge); after dynamic adjustment fails, current fluctuations continue to exceed [a certain threshold]. (If a latent fault exists in the power supply circuit). At this time, the protection control unit immediately performs the following actions: sends a shutdown signal to the power switch (MOSFET), cutting off the power supply circuit from the Type-C interface to the internal module of the device. The power-off response time is ≤10μs to avoid prolonged impact of large current on the module; triggers the fault lockout mechanism: marks the protection status as "overcurrent lockout" and sends a "device overcurrent protection" fault code (PD protocol fault code is 0x05, QC protocol fault code is 0x08) to the mobile phone via the Type-C interface. Even if the Type-C interface is unplugged and plugged back in, the device will not automatically restore power. The user must send a "fault reset" command via the accompanying mobile app for the protection control unit to clear the lockout status; records fault information: records the fault information at the time of the overcurrent occurrence. , Parameters such as duration and trigger level are stored in the device's non-volatile memory (NVM) for easy access to fault logs via the APP to locate the cause of the fault (such as module short circuit, Type-C interface damage, etc.).

[0039] When the first / second level of protection is triggered, the power supply remains uninterrupted. The UAC protocol parsing of the mobile phone audio conversion module and the ADC sampling of the microphone audio conversion module can proceed normally. Only the currently processed signal frames (such as UAC protocol probe frames and microphone analog signals) need to be buffered internally within the modules to prevent signal loss due to current fluctuations. If the third level of forced power failure is triggered, the protection control unit will send a "power failure warning" signal to both core modules (20 seconds in advance) before the power failure. Upon receiving a signal, the module immediately stops processing the current signal and saves the processed intermediate data (such as audio conversion configuration parameters and digital radio signals). / / Once the fault is reset and power is restored, the system can continue working directly based on the saved data without restarting the adaptation process, thus reducing initialization time.

[0040] While the power processing module provides stable power, the mobile audio conversion module establishes data interaction with the mobile system through the D+ and D- data cables of the Type-C interface, and prioritizes parsing the USB audio underlying protocol (UAC, Universal Serial Bus Audio Class) that is uniformly followed by all Android systems.

[0041] The module first sends a 64-byte, 115200bps protocol probe frame and reads the UAC protocol version information returned by the mobile phone: if the "bcdADC" field value in the protocol is 0x0100, it is determined to be the UAC1.0 protocol; if it is 0x0200, it is determined to be the UAC2.0 protocol. Regardless of the protocol version, it extracts the current audio output mode (digital signal "D" or analog signal "A") and the original sampling rate from the "audio configuration descriptor" field of the UAC protocol. (44100Hz or 48000Hz) and the original bit depth ( (16-bit or 24-bit) as the basic parameters for subsequent signal processing.

[0042] When identified as a digital PCM signal, the module calls the general PCM decoding module for normalization processing: if the original sampling rate... Original depth The sampling rate is increased to 48000Hz using a linear interpolation algorithm, and the bit depth is extended to 24 bits by padding high-order bits with zeros; if the original parameters are already... , If the parameters are not specified, they are retained. The final generated digital audio stream conforms to the Bluetooth transmission module input standard, and its format is defined by the formula: .

[0043] When identified as an analog AC coupled signal, the processing flow consists of three steps: First, the original analog signal is sampled using a 16-bit resolution ADC (Analog-to-Digital Converter) to obtain the original amplitude of the analog signal. Range -1.5V (+1.5V); Second, activate the Automatic Gain Control (AGC) module to normalize the original amplitude to -1V. The +1V range is specifically implemented using the gain formula: (in This is the gain coefficient, with a value ranging from 0.67. 1.0, for example At 1.5V, =0.67, calculated as follows =1V; At 0.8V, =1.0, calculated as follows =0.8V); the third step is to convert the normalized analog signal back into a digital signal via an ADC, setting the sampling rate to 44100Hz and the bit depth to 16bit. The final standard audio stream format is defined by the formula: .

[0044] After completing signal processing, the module generates a signal processing mode (digital decoding / AGC+ADC) and a standard sampling rate based on the aforementioned standard audio stream. Standard depth The audio conversion configuration parameters are transmitted to the microphone audio conversion module specified in the handover document, providing an adaptation basis for the unified processing of subsequent audio signals.

[0045] While power supply and audio adaptation are being promoted simultaneously, the bidirectional data channel of the Type-C interface (following the USBHID1.1 protocol) initiates scene mode command reception to ensure that the device's audio mode matches the user's actual usage scenario.

[0046] First, the accompanying mobile app generates corresponding scene mode commands based on the user's selected scenario: when the user selects "Meeting Mode," command "M1" is generated; when "Outdoor Mode" is selected, command "M2" is generated. To avoid misjudgment due to interference during command transmission, the app calculates the CRC checksum of the command using a specific polynomial. The generator polynomial is The scene mode command and CRC checksum are encapsulated into a 16-byte command data frame, and the encapsulation format is defined by the formula: (in (This is a scene mode instruction, with values ​​"M1" or "M2")

[0047] After the instruction data frame is transmitted to the device via the D+ and D- data lines of the Type-C interface, the instruction receiving module on the device side first extracts the CRC checksum from the frame. ), and then the scene mode command ( Recalculate the checksum. ), through comparison and Verify instruction integrity: If the two are equal, the instruction is considered to have been received accurately; if they are not equal, immediately send a retransmission request to the mobile APP until the instruction that has passed verification is received.

[0048] After successful verification, the module extracts the scene mode command from the command data frame. The data is stored in the device's temporary parameter cache and synchronized in real time to the microphone audio conversion module. At this time, the stable power supply from the power processing module ensures the microphone module can respond normally to commands, and the audio conversion configuration parameters of the mobile phone audio conversion module provide a standard for signal processing after microphone pickup. These three components work together to... Precisely trigger the switching between "mobile phone microphone channel / device built-in microphone channel" to prepare for the subsequent audio control steps in the handover document.

[0049] For example, after the device and mobile phone are physically connected via a Type-C interface, the power processing module sends a 1kHz sine wave probe signal to the mobile phone through the CC line (ConfigurationChannel). The signal amplitude is 3.3V, and the duration is 100ms. After receiving the probe signal, the mobile phone sends back a response frame containing the "QC_ACK" identifier (frame format: 0x550x010x04 "QC_ACK" 0xAA). The device parses the frame and determines that the current power supply protocol is QC (QuickCharge). It then outputs the standard power supply voltage according to the QC protocol. Safe power supply current Calculated according to the formula The total power consumption of the core modules = power consumption of the mobile phone audio conversion module (3W) + power consumption of the microphone audio conversion module (2W) = 5W. This meets the power supply requirements. The power processing module acquires real-time current at a sampling frequency of 10kHz through a series 0.01Ω sampling resistor (accuracy 0.1%). The mean value of 100 consecutive sampling data in the experiment was 1.99A. The result was calculated using the formula... , The current was determined to be stable, with no protection action detected. The mobile phone audio conversion module sends a 64-byte protocol probe frame to the mobile phone via the Type-C D+ and D- data lines at a baud rate of 115200bps. Example frame content: 0x000x010x02...0x3F (64 bytes in total, including the UAC protocol probe identifier).

[0050] The phone returns a UAC protocol frame. Parsing the "bcdADC" field reveals a value of 0x0200, which, according to the documentation, indicates a UAC 2.0 protocol. Further extraction of the "Audio Configuration Descriptor" field reveals the audio output mode: digital signal (identified by "D"); original sampling rate. Original depth .because Initiate the linear interpolation algorithm. Interpolation factor. Select Adjacent sampling points in (original digital signal) (time ), (time , ), calculate target sampling points ( interpolation coefficients Calculate the target sample value using the linear interpolation formula. Ultimately, the sampling rate will be increased to .because The bit depth is extended to 24 bits by padding with eight zeros at the high bits. Example: 16-bit data 0x1234 → 24-bit data 0x001234. This generates a digital standard audio stream. The original amplitude was obtained by sampling the analog signal using a 16-bit ADC. (at -1.5V) +1.5V range). According to the formula ,because ,Pick (make ). Calculated Satisfying "-1V" The +1V normalization requirement is applied. The normalized analog signal is converted into a digital signal (44100Hz / 16bit) via an ADC, and then generated according to the formula in the document. Generates a signal processing mode including "digital decoding, , The configuration parameters are transmitted to the microphone audio conversion module via the internal bus, providing a standard for subsequent audio conversion.

[0051] Select "Meeting Mode" on the mobile app and generate instructions. "M1". Generate the polynomial based on the document's CRC. Calculate the CRC checksum for "M1" (ASCII code: 0x4D0x31): Initial CRC value = 0xFF; After byte-by-byte XOR calculation, we get... According to the formula Encapsulated as a 16-byte data frame, example 0x4D0x310x000x00...0x7A (total 16 bytes, "M1" occupies the first 2 bytes), Occupying the 16th byte (padded with 0x00 in the middle). The device receives a 16-byte data frame via the Type-C D+ and D- data lines, and extracts... Separation command "M1". Recalculate the CRC for "M1". Based on logical comparison and : If the instruction was received accurately and there was no retransmission request, then... The device will immediately send a retransmission request to the APP (frame format: 0xEE0x01 "RETRY" 0xFF). The verified "M1" command will be stored in a temporary parameter buffer and synchronized to the microphone audio conversion module; at this time, the power processing module will continuously output 9V / 2A. (Stable), the mobile phone audio conversion module has output configuration parameters (48kHz / 24bit), and the three work together to trigger the dual-channel start of the "mobile phone microphone" and "device built-in microphone", preparing for the subsequent sound recording control stage.

[0052] Based on the scene mode commands (M1 / M2) obtained above, the system implements differentiated sound pickup control for the dual microphone channels: when the "Conference Mode" command M1 is received, the activation signals of the mobile phone microphone and the device's built-in microphone are triggered simultaneously, and the mobile phone microphone acquires the original analog signal of ambient sound at a sampling rate of 44.1kHz and a bit depth of 16bit. Amplitude range -1V +1V, and transmits it to the microphone audio conversion module via the Type-C interface; the device's built-in microphone (sensitivity -38dBV / Pa) acquires the original analog signal of the target speech at a sampling rate of 48kHz and a bit depth of 24bit. Amplitude range -0.8V +0.8V) is directly sent to the microphone audio conversion module. When the "Outdoor Mode" command M2 is received, the system sends a shutdown signal to the phone's microphone, keeping only the device's built-in microphone active and activating its wind noise reduction hardware circuit to acquire the original analog wind noise reduction signal ( Amplitude range -0.6V +0.6V) is transmitted to the microphone audio conversion module. Subsequently, the analog-to-digital converter (ADC) within the microphone audio conversion module converts the received signal... , or Converted into digital audio signals with source identifiers ("mobile phone microphone" / "device microphone") respectively. : 44.1kHz / 16bit; : 48kHz / 24bit; (48kHz / 24bit).

[0053] Specifically, scene mode instructions are completed on the Type-C bidirectional data channel. After receiving and verifying the data, the system, based on the hardware architecture of "microphone audio conversion module + dual microphone channels" and combined with the stable power supply output from the previously processed power module, , The audio conversion configuration parameters (including standard sampling rate) generated by the mobile phone audio conversion module are the same as those generated by the mobile phone audio conversion module. Standard depth Differentiated sound pickup control is implemented for the dual microphone channels to ensure that the acquired signals not only meet the subsequent processing standards but also accurately match the user scenario. The specific process is as follows:

[0054] First, the system retrieves the validated scene mode instructions from the temporary parameter cache. This command serves as the core trigger signal for dual-microphone channel switching. (In "Conference Mode"), the system simultaneously sends a start signal to both the phone's microphone and the device's built-in microphone, and assigns acquisition parameters adapted to the previously configured audio conversion parameters to both microphones: the phone's microphone starts acquisition at a 44.1kHz sampling rate and 16-bit bit depth. These parameters are consistent with the standard sampling rate (44100Hz) and standard bit depth (16-bit) when the phone's audio conversion module processes analog signals, ensuring seamless processing of subsequent signals. The acquired ambient sound is the original analog signal. The amplitude is strictly controlled at -1V. To avoid signal overload and distortion within the +1V range, after acquisition, the data is transmitted to the microphone audio conversion module via the auxiliary data channel of the Type-C interface; the device's built-in microphone (whose sensitivity is specified in the manual as -38dBV / Pa) starts acquisition at a 48kHz sampling rate and 24-bit bit depth. These parameters match the standard sampling rate (48000Hz) and standard bit depth (24bit) of the mobile phone's audio conversion module when processing digital signals. The target audio signal acquired is the original analog voice signal (…). Amplitude controlled at -0.8V +0.8V ensures sufficient signal dynamic range while reducing background noise interference. The signal is directly transmitted to the microphone audio conversion module through the internal circuitry of the device, enabling dual-source collaborative acquisition of "ambient sound" and "target sound". This meets the needs of accurately capturing target speech and simultaneously perceiving the ambient atmosphere in meeting scenarios.

[0055] when (In "Outdoor Mode"), the system sends a shutdown signal to the phone's microphone to prevent noise from mixing in due to its weak wind noise resistance in outdoor environments. Simultaneously, it only sends an activation signal to the device's built-in microphone and uses an adaptive filtering algorithm for noise reduction: At this time, the device's built-in microphone still collects data at a 48kHz sampling rate and 24-bit bit depth, but the adaptive filtering algorithm pre-filters low-frequency wind noise below 500Hz, resulting in a better wind noise reduction original analog signal. Amplitude controlled at -0.6V +0.6V further compresses the noise amplitude range, and the signal is directly transmitted to the microphone audio conversion module to ensure that the signal collected in outdoor scenarios is mainly the target voice and reduce environmental wind noise interference.

[0056] The collected original analog signal for wind noise reduction ( The input consists of an adaptive filtering algorithm composed of time-domain adaptive filtering and frequency-domain notch filtering, which is used to reduce wind noise in the original analog signal ( Before being transmitted to the ADC for conversion, the audio is first denoised by the digital signal processor (DSP) built into the microphone audio conversion module, and then enters the subsequent analog-to-digital conversion process.

[0057] Device built-in microphone for data acquisition Then, the DSP first presamples the analog signal (sampling frequency 48kHz) to generate a temporary digital signal. An adaptive filtering model was constructed based on the "time-domain characteristics of wind noise signals (low frequency, stable, no obvious pulses)". Extraction was performed using a sliding window (window length 50ms). Low-frequency components below 500Hz are used as wind noise reference signals. (formula: ,in This is a Butterworth low-pass filter (4th order). The Least Mean Square (LMS) algorithm is used to... For reference, Filtering is performed using the formula Dynamically adjust the filter coefficient ( for Time coefficient vector Step size factor For error signals, (Input signal vector), initially suppressing steady wind noise below 500Hz, output time-domain filtered signal. .

[0058] right A Fast Fourier Transform (FFT) is performed to transform the signal to the frequency domain, followed by notch filtering to further remove low-frequency wind noise that was not fully suppressed by the time-domain filtering. The power spectral density (PSD) of the frequency domain signal is analyzed to pinpoint the frequency band where the power proportion below 500Hz exceeds 30% (typically 100Hz). 500Hz), marked as the core frequency band for wind noise. .against Construct a multi-frequency notch filter using the formula ( The notch frequency, Let be the radius of the pole. The attenuation coefficient is... (Number of notch points) The wind noise frequency band was analyzed using a 10-fold method. 15dB attenuation, output frequency domain filtered signal .right Perform an inverse fast Fourier transform (IFFT) to restore the signal to the time domain, obtaining the original analog signal after denoising and wind noise reduction. The amplitude range remains controlled within -0.6V. +0.6V).

[0059] After acquiring the raw analog signals in different scenarios, the microphone audio conversion module initiates the analog-to-digital conversion process. The ADC (analog-to-digital converter) parameters within the module are precisely matched with the acquisition parameters of the two microphones: for the raw analog signals of ambient sound ( The ADC converts the signal at a sampling rate of 44.1kHz and a resolution of 16bit to generate a digital audio signal with a "mobile phone microphone" source identifier. Its conversion parameters (sampling rate, bit depth) must be consistent with the standard audio stream, and the format is defined as follows: ;Targeting the original analog signal of the target speech ( ) and the original analog signal after noise reduction (for wind noise) The ADCs were all converted at a sampling rate of 48kHz and a resolution of 24bit, generating digital audio signals with a "device microphone" source identifier. ) and digital radio signals ( The format definitions are as follows: , The converted digital radio signal ( / / The audio will be directly transmitted to the AI ​​audio processing module, and compared with the standard digital audio stream previously generated by the mobile phone audio conversion module. Collaborative processing ensures consistent parameters and logical coherence throughout the entire process.

[0060] For example, the system retrieves validated instructions from the temporary parameter buffer. "M1" indicates a meeting mode has been selected, triggering the dual-channel activation of both the "mobile phone microphone" and the "device's built-in microphone." Based on the configuration parameters of the mobile phone's audio conversion module, acquisition parameters are assigned to the two microphones, and the mobile phone microphone is matched to the analog signal standard. , Amplitude limit -1V +1V; The device's built-in microphone is compatible with digital signal standards. , Amplitude limit -0.8V +0.8V. The mobile phone microphone was used to capture conversation (ambient noise) 3m behind in the conference room, with a sound pressure level of 55dB, generating the original analog signal of the ambient sound. Amplitude range 0.3V 0.5V (≤1V, no overload), transmitted to the microphone audio conversion module via the Type-C auxiliary data channel. The device's built-in microphone captures the speech (target signal) of a speaker 1m directly in front of the conference room, with a sound pressure level of 70dB; generating the original analog signal of the target speech. Amplitude range 0.4V 0.7V (≤0.8V, dynamic range preserved) is transmitted to the microphone audio conversion module through the device's internal circuitry.

[0061] System Extraction Command "M2" sends a mute signal to the phone's microphone (to prevent wind noise from mixing in), activating only the device's built-in microphone. The data acquisition scenario is an outdoor wind speed of 3 m / s (primarily low-frequency wind noise, with 35% power allocated to frequencies below 500 Hz), and the target speech is a speaker 2 meters ahead (sound pressure level 65 dB); generating a wind-noise-reduced raw analog signal. Amplitude range 0.2V 0.4V (≤0.6V, compressed noise amplitude) is transmitted to the DSP unit of the microphone audio conversion module. The DSP... according to Presampling generates temporary digital signals Through a 4th-order Butterworth low-pass filter ( According to the formula Low-frequency components below 500Hz were extracted as wind noise reference signals. (Amplitude 0.1V) 0.3V). Let the step size factor be... , Time coefficient vector (16 orders in total), input signal vector Error signal Update coefficients according to the formula Time-domain filtering output: After 300 iterations, a time-domain filtered signal is generated. Wind noise attenuation is 8dB below 500Hz. Perform a 2048-point FFT (consistent with the FFT point count in the AI ​​module) to analyze the power spectral density (PSD) and pinpoint the core frequency band of wind noise. .set up , , notch point (Corresponding to 150Hz, 300Hz, and 450Hz), the notch frequency is calculated. , , .right The frequency band implements 12dB attenuation (total attenuation 8dB + 12dB = 20dB), output Perform IFFT to restore the signal to the time domain, and obtain the denoised signal. The amplitude is still controlled at 0.2V. 0.4V (≤0.6V). ADC conversion according to "44.1kHz / 16bit" parameters (matching mobile phone audio analog standards), based on format definition. The amplitude range of the converted digital signal corresponds to 0.3V. 0.5V (quantization error ≤ 0.001V). ADC converts according to "48kHz / 24bit" parameters (matching mobile phone audio digital standards), based on the format definition. The amplitude range of the converted digital signal corresponds to 0.4V. 0.7V (quantization error ≤ 0.0001V), preserving target speech details. The ADC converts the denoised speech using the "48kHz / 24bit" parameters. According to the format definition After conversion, the proportion of wind noise power below 500Hz in the digital signal is reduced to 5%, and the target speech signal-to-noise ratio is improved to 58dB.

[0062] The AI ​​audio processing module receives the digital audio signal output by the microphone audio conversion module. / / After obtaining the source identifier, targeted processing will be carried out: if the input contains identifiers for "mobile phone microphone" and "device microphone"... and The module first calculates the time-domain delay of the two signals. (Accuracy 1ms) — Through comparison and By analyzing the time difference of the same sound peak, the direction of the target sound can be accurately located; then, the frequency domain characteristic difference between the two signals is analyzed. (Energy percentage difference in the 1kHz-3kHz frequency band, accuracy 0.1%), effectively distinguishing speech from environmental noise. Then, using... As the core signal, combined A filtering model is constructed based on the noise characteristics to remove noise that is... Interference signals with noise characteristics (such as conversations in the back of a conference room) are ultimately used to generate enhanced digital audio. (Sampling rate 48kHz / 24bit, signal-to-noise ratio ≥60dB). If the input is labeled "Device Microphone". The module activates the anti-wind noise algorithm to... The wind noise frequency band below 500Hz is attenuated by 10-15dB, while the speech frequency band from 300Hz to 3kHz is gained by 3-5dB, thus generating enhanced digital audio.

[0063] Specifically, when the input signal is identified by both "phone microphone" and "device microphone" identifiers... (44.1kHz / 16bit) and At (48kHz / 24bit) (corresponding to conference mode M1), the module first activates the multi-source signal synchronization mechanism, using an interpolation algorithm to synchronize the signals. The sampling rate has been increased to 48kHz, and The sampling rate is unified, laying the foundation for subsequent time-domain and frequency-domain analysis. The first step is to calculate the time-domain delay of the two signals. : Traverse via a sliding window (window length 20ms) and Locate the peak point of the same sound event (such as a speaker's voice) according to the formula. Calculate the time difference (with precision controlled within 1ms), where for The time of occurrence of the mid-peak for The timing of the same peak occurrence is used to pinpoint the direction of the target sound. A positive value indicates that the microphone picked up the sound first, while a negative value indicates the opposite. The second step involves analyzing the frequency domain characteristic differences. : After synchronization and Perform Fourier transform to extract the energy percentage of the core audio segment from 1kHz to 3kHz. for The proportion, for (the percentage), according to the formula Calculate the difference (accuracy 0.1%) – typically the target speech is in The proportion of middle school students is higher ( ), while environmental noise (such as conversations in the back row) The proportion of middle school students is higher ( This allows for the initial distinction between speech and noise.

[0064] The specific calculation of the energy percentage is as follows: For example, first... Perform frame segmentation, maintaining a frame length of 20ms. Calculate the number of sampling points per frame point( (where the frame length is 1). A Hanning window is applied to each frame of signal, and the window function formula is: After windowing, the signal is It is used to reduce spectral distortion caused by signal abrupt changes at frame edges.

[0065] The radix-2 fast Fourier transform algorithm is used to process the windowed single-frame signal. The transformation is performed as follows: if the frame length ,right Perform zero-padding (add 1088 zeros) to make the total length... Ensure the frequency resolution meets the requirements. Perform an FFT transform to convert the time-domain signal... Convert to frequency domain signal The formula is ,in, This is a frequency index, corresponding to the actual frequency. (For example hour, (i.e., the starting point of the core language frequency band).

[0066] Through frequency domain signals Since it is a complex number, its power value is the square of the modulus divided by the number of FFT points, as shown in the formula: (Right now for Real part, for (Imaginary part). Based on actual frequency. ,Sure and Corresponding frequency index range Calculation formula , ( To round up, To round down and ensure full coverage of the 1kHz-3kHz frequency band. Total power of the core audio band. Total power across the entire frequency band (0 24000Hz, that is to ) (Due to FFT symmetry, only the first half needs to be calculated). The energy percentage formula is: And so on, to obtain...

[0067] Based on the above analysis, the module is... (The target voice signal collected by the device's microphone) is the core, combined with An adaptive filtering model is constructed based on the noise characteristics: by... middle The frequency band is marked as an "environmental noise template" for... The signal matched to this template is dynamically attenuated (5-10dB), with a focus on eliminating interference such as conversations in the back of the meeting room. The final generated enhanced digital audio... Acquisition process: For noise-filtered Perform amplitude normalization (adjust the signal amplitude to -0.5V). (within the +0.5V range), then through 24-bit quantization and deep resampling, the final generated format is unified as follows. Enhanced digital audio, where the signal-to-noise ratio (SNR) is calculated according to the formula calculate( For target voice power, (Residual noise power).

[0068] The adaptive filtering model construction process is as follows: By middle The frequency bands are labeled as "noise sub-bands," and the power values ​​of all noise sub-bands are extracted. Constructing a "noise template power spectrum" As a follow-up The benchmark for interference matching. (For) The 1kHz-3kHz frequency band sub-band, combined with an environmental noise template, is used to filter interference signals. Each sub-band Calculate its power Power spectrum of noise template The similarity is calculated using the formula: (For example , ,but The higher the similarity, the more likely the sub-band is to be an interference signal. And this sub-band belongs to the marked "noise sub-band" ( If it is, then it is determined to be an "interference subband". The dual judgment criteria are adopted. The noise discrimination logic further excludes noise based on power similarity. Noise subband but Misjudgment due to "language". Calculate based on the similarity and power difference of the interference subbands. The suppression gain of the interference subband is set to ensure that the attenuation meets the requirement of "5-10dB", as shown in the following formula. ,when , When the interference intensity is consistent with the noise template, It is within the "5-10dB" attenuation range; when , When the interference intensity is twice that of the template, If the value exceeds the limit, it will be truncated by the upper limit of 10dB to ensure compliance with the constraint of "maximum 10dB attenuation"; non-interference subband ( )of Completely preserved The target speech signal is processed to avoid damaging useful information. Suppression gain is applied to... The frequency domain signal is reconstructed into a time domain signal. (The process involves) obtaining... Frequency domain signal after Fourier transform For each interference sub-band corresponding frequency point Adjust the gain according to the suppression, the formula is as follows: ( for The original frequency domain amplitude is preserved, while retaining the original phase information to avoid signal distortion. This applies to the suppressed frequency domain signal. Perform a 2048-point IFFT to convert it into a time-domain signal. Remove the IFFT zero-padding, retain the time-domain signal with the same length as the original frame (960 points, 20ms), and then concatenate all frames using an overlap-addition method (50% overlap rate) to generate the filtered signal. (48kHz / 24bit). For Perform amplitude normalization (adjust to -0.5V). +0.5V), ultimately generating enhanced digital audio that conforms to the document standards. It is directly transmitted to the Bluetooth transmission module.

[0069] When the input signal is a single identifier for "Device Microphone" At (48kHz / 24bit) (corresponding to outdoor mode M2), the module directly activates the wind noise reduction algorithm: first, it processes the low-frequency wind noise band below 500Hz using a Butterworth high-pass filter, according to the formula... Calculate the attenuation ( The cutoff frequency, (For signal frequency), ensure attenuation of 10dB at 500Hz and 15dB at 250Hz; simultaneously activate dynamic gain control for the 300Hz-3kHz speech band, according to the formula... Adjusting the gain ( (The signal frequency is used to set a gain of 3dB at 300Hz and a gain of 5dB at 3kHz, compensating for speech attenuation while avoiding excessive amplification of high-frequency noise. The final generated enhanced digital audio format is...) It maintains compatibility with the signal parameters output in conference mode, preparing for subsequent Bluetooth transmission to the hearing aid.

[0070] After receiving the enhanced digital audio output from the AI ​​audio processing module, the Bluetooth transmission module reads the connection status parameters (stuttering frequency f: times / minute; transmission rate v: Mbps) from the mobile phone in real time via the Type-C interface: When the stuttering frequency f > 2 times / minute or the transmission rate v < 2 Mbps, the module automatically switches the Bluetooth transmission protocol from APT-X to SBC, encapsulating the enhanced digital audio into an SBC format Bluetooth signal according to the SBC encoding format (bitrate 320kbps, stereo channel mode); when the stuttering frequency f ≤ 2 times / minute and the transmission rate v ≥ 2 Mbps, the APT-X protocol is maintained, and the S4 signal is encapsulated into an APT-X format Bluetooth signal according to the APT-X encoding format (bitrate 576kbps, stereo channel mode). Finally, the SBC format Bluetooth signal or the APT-X format Bluetooth signal is wirelessly transmitted to the Bluetooth hearing aid via the 2.4GHz band, where it is decoded into an analog audio signal (amplitude range -1V). +1V) drives the earpiece to play, with the latency controlled within 100ms throughout to avoid audio-visual desynchronization issues.

[0071] Specifically, the Bluetooth transmission module receives enhanced digital audio output from the AI ​​audio processing module. or After that, the connection status parameters (lag frequency) fed back by the mobile phone are read in real time through the Type-C interface. : times / minute; transmission rate :Mbps):

[0072] When stuttering frequency is detected times / minute or transmission rate At Mbps, the module automatically switches the Bluetooth transmission protocol from APT-X to SBC. If the input is enhanced digital audio in conference mode... Then Encapsulated into SBC format Bluetooth signals according to SBC encoding format (bit rate 320kbps, stereo channel mode). If the input is outdoor mode enhanced digital audio Then Encapsulated into SBC format Bluetooth signals using the same SBC encoding format. (The coding process uses formulas) Dynamically adjust quantization accuracy , The larger (Appropriately reduce to improve packet loss resistance).

[0073] When stuttering frequency times / minute and transmission rate At Mbps, maintain the APT-X protocol — if the input is ,Will The signal is encapsulated in APT-X format (576kbps bitrate, stereo channel mode) as a Bluetooth signal. If the input is ,Will Bluetooth signals are encapsulated in the same APT-X encoding format and then packaged into APT-X format Bluetooth signals. .

[0074] Finally, the Bluetooth signal in the corresponding mode ( , , or It employs Frequency Hopping Spread Spectrum (FHSS) technology to dynamically switch channels within the 2.4 GHz band (2402-2480 MHz) at a hopping rate of 1600 times / second, avoiding interference from devices such as Wi-Fi, and transmits the signal to the Bluetooth hearing aid. The hearing aid then decodes the signal into an analog audio signal (amplitude range -1V). +1V) drives the earpiece for playback, and the latency is kept below 100ms throughout the process through protocol optimization to avoid audio-visual desynchronization issues. Each frame of data is appended with a 3-byte CRC checksum (generator polynomial). ), to ensure transmission integrity.

[0075] This process uses contextualized signal identification ( / ) and Bluetooth signal format ( / The one-to-one correspondence ensures seamless connection with the output parameters of the AI ​​audio processing module mentioned above, forming a complete signal transmission link.

[0076] After receiving the signal, the hearing aid's decoding module automatically matches the decoder (SBC decoder or APT-X decoder) according to the signal format, restoring it to a digital audio signal; then, a digital-to-analog converter (DAC) converts the digital signal into analog audio (amplitude range -1V). +1V), conversion accuracy of 16-bit, sampling rate of 48kHz; finally, the analog audio is converted into sound wave output through the earpiece, and the end-to-end latency is controlled within 80-100ms (protocol switching time ≤10ms), meeting the needs of real-time communication.

[0077] The processes described above with reference to the flowcharts in the embodiments disclosed in this invention can be implemented as computer software programs. The embodiments disclosed in this invention include a computer program product comprising a computer program carried on a computer-readable medium, the computer program containing program code for performing the methods shown in the flowcharts. In such embodiments, the computer program can be downloaded and installed from a network via a communication component, and / or installed from a removable medium. When the computer program is executed by a central processing unit (CPU), it performs the functions defined in the methods of this application. It should be noted that the computer-readable medium described above in this application can be a computer-readable signal medium or a computer-readable storage medium, or any combination of the two. A computer-readable storage medium can be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any combination thereof. More specific examples of computer-readable storage media may include, but are not limited to: electrical connections having one or more wire segments, portable computer disks, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fiber, portable compact disk read-only memory (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination thereof. In this application, a computer-readable storage medium can be any tangible medium containing or storing a program that can be used by or in connection with an instruction execution system, apparatus, or device. In this application, a computer-readable signal medium may include a data signal propagated in baseband or as part of a carrier wave, carrying computer-readable program code. Such propagated data signals can take various forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination thereof. A computer-readable signal medium can also be any computer-readable medium other than a computer-readable storage medium, which can send, propagate, or transmit a program for use by or in connection with an instruction execution system, apparatus, or device. The program code contained on a computer-readable medium may be transmitted using any suitable medium, including but not limited to: wireless segments, wire segments, optical fibers, RF, etc., or any suitable combination thereof.

[0078] The flowcharts and block diagrams in the accompanying drawings illustrate the architecture, functionality, and operation of possible implementations of systems, methods, and computer program products according to various embodiments of the present invention. In this regard, each block in a flowchart or block diagram may represent a module, segment, or portion of code containing one or more executable instructions for implementing a specified logical function. It should also be noted that in some alternative implementations, the functions indicated in the blocks may occur in a different order than those indicated in the drawings. For example, two consecutively indicated blocks may actually be executed substantially in parallel, and they may sometimes be executed in reverse order, depending on the functions involved. It should also be noted that each block in the block diagrams and / or flowcharts, and combinations of blocks in the block diagrams and / or flowcharts, can be implemented using a dedicated hardware-based system that performs the specified function or operation, or using a combination of dedicated hardware and computer instructions.

[0079] Those skilled in the art should understand that the embodiments of the present invention described above and shown in the accompanying drawings are merely examples and do not limit the present invention. The purpose of the present invention has been fully and effectively achieved. The functions and structural principles of the present invention have been shown and explained in the embodiments. Without departing from the stated principles, the implementation of the present invention may have any variations or modifications.

Claims

1. A method for enhancing the directivity and clarity of audio in a dual-channel hearing aid, characterized in that, The method includes: The power processing module sends a detection signal to identify the mobile phone power supply protocol. The mobile phone audio conversion module receives and identifies the UAC protocol. It outputs the corresponding safe voltage and current according to the type of mobile phone power supply protocol, determines whether the power supply meets the requirements of the core module, monitors the current in real time, and triggers three-level protection based on the current deviation. It identifies the current audio output mode, sampling rate and bit depth of the mobile phone by parsing the UAC protocol, and uses a general adaptation algorithm to normalize and obtain the standard audio stream format. Receive scene mode instructions, control the dual microphone channels to acquire analog signals according to the scene mode instructions, and convert the acquired analog signals into digital radio signals with source identification through analog-to-digital conversion; Determine the type of digital audio signal. If it is a dual-identifier signal, calculate the time-domain delay of the two signals, analyze the frequency domain feature difference to distinguish between speech and environmental noise, and construct a filtering model based on the device microphone signal and the noise characteristics of the mobile phone microphone signal to remove interference signals and obtain enhanced digital audio. If it is a single-identifier signal, generate enhanced digital audio through an anti-wind noise algorithm. The system reads the connection status parameters fed back by the mobile phone in real time, and encapsulates the enhanced digital audio into Bluetooth signals in different ways according to the connection status parameters and transmits them to the Bluetooth hearing aid. The specific process of obtaining the digital radio signal is as follows: The scene mode command that has passed CRC verification is extracted from the temporary parameter buffer. The scene mode command type is determined. If it is a conference mode, the mobile phone microphone and the device's built-in microphone are activated simultaneously. The mobile phone microphone collects the original analog signal of ambient sound at a sampling rate of 44.1kHz and a bit depth of 16 bits, while the device's built-in microphone collects the original analog signal of target voice at a sampling rate of 48kHz and a bit depth of 24 bits. If it is an outdoor mode, a shutdown signal is sent to the mobile phone microphone, and only the device's built-in microphone is activated. This microphone collects at a sampling rate of 48kHz and a bit depth of 24 bits, and pre-filters low-frequency wind noise below 500Hz using an adaptive filtering algorithm to generate a wind noise-reduced original analog signal. The microphone audio conversion module starts analog-to-digital conversion. The ADC parameters in the module are precisely matched with the corresponding microphone acquisition parameters. For the original analog signal of ambient sound collected by the mobile phone microphone, the ADC is converted at a sampling rate of 44.1kHz and a resolution of 16 bits to generate a digital audio signal with the mobile phone microphone source identifier. For the original analog signal of target voice and the wind noise-reduced original analog signal collected by the device's built-in microphone, the ADC is converted at a sampling rate of 48kHz and a resolution of 24 bits to generate digital audio signals with the device microphone source identifier.

2. The method for enhancing the directivity and clarity of audio in a dual-channel hearing aid according to claim 1, characterized in that, The specific process for determining whether the power supply meets the requirements of the core module is as follows: The power processing module is physically connected to the mobile phone via the Type-C interface. It sends a detection signal through the CC line to identify the mobile phone's power supply protocol. Based on the identified mobile phone power supply protocol, it outputs the corresponding standard power supply voltage and safe power supply current. If it is the PD protocol, it outputs 5V voltage and 2A current. If it is the QC protocol, it outputs 9V voltage and 2A current. The power processing module calculates the actual power supply by multiplying the output voltage and current according to the power supply calculation formula. The calculated actual power supply is compared with the total power consumption of the core module of the device to determine whether the current power supply can meet the operating requirements of the core module.

3. The method for enhancing the directivity and clarity of audio in a dual-channel hearing aid according to claim 2, characterized in that, The specific details of the three-level protection are as follows: The power processing module monitors the supply current in real time and calculates the current deviation. The three-level protection is triggered based on the current deviation and its duration; when When this occurs, the first level of protection is triggered, sending a current fluctuation warning signal to the mobile phone via the Type-C interface, and caching the warning log internally within the device; if And lasting for more than 50 This triggers the second level of protection. The protection control unit sends a command to the power management chip, which fine-tunes the output voltage using PWM technology and reverses the current to return it to a stable range, every 10 seconds. If the adjustment parameters are updated and fail to work after three consecutive attempts, the third level of protection will be triggered; when And the duration exceeds 50 The third-level protection is triggered, the power supply circuit is immediately cut off, the fault lockout is triggered, and the current, voltage and duration of the overcurrent are stored in the non-volatile memory. Before the power is cut off, a warning signal is sent to the core module to save the intermediate data.

4. The method for enhancing the directivity and clarity of audio in a dual-channel hearing aid according to claim 3, characterized in that, The specific process for obtaining the standard audio stream format is as follows: If identified as a digital PCM signal, the general PCM decoding module performs standardization processing. If the original sampling rate is 44100Hz and the original bit depth is 16bit, the sampling rate is increased to 48000Hz using a linear interpolation algorithm, and the bit depth is extended to 24bit by padding high-bits with zeros. If the original parameters are already 48000Hz and 24bit, the parameters are directly retained, generating a standard digital audio stream format. If identified as an analog AC coupled signal, the original analog signal is sampled using a 16-bit resolution ADC to obtain the original amplitude. The automatic gain control module is then activated to normalize the original amplitude to -1V. In the +1V range, the normalized analog signal is converted into a digital signal via an ADC, with a sampling rate of 44100Hz and a bit depth of 16bit, generating a standard analog-to-digital audio stream format.

5. The method for enhancing the directivity and clarity of audio in a dual-channel hearing aid according to claim 4, characterized in that, The enhanced digital audio acquisition process is as follows: If the input is a digital audio signal with dual identifiers for both the mobile phone microphone and the device microphone, the sampling rate of the mobile phone microphone signal is increased to 48kHz through an interpolation algorithm. The time-domain delay of the two signals is calculated to locate the direction of the target sound. The frequency domain feature difference in the 1kHz-3kHz band is analyzed to distinguish speech from environmental noise. Taking the device microphone signal as the core, an adaptive filtering model is constructed by combining the noise feature of the mobile phone microphone signal with a frequency domain feature difference of less than 5%. The interference signal in the device microphone signal that matches the noise template is dynamically attenuated. After amplitude normalization and 24-bit resampling, enhanced digital audio is generated. If the input is a digital audio signal with only a single identifier for the device microphone, the wind noise reduction algorithm is directly activated. The Butterworth high-pass filter attenuates the low-frequency wind noise band below 500Hz by 10-15dB and dynamically gains the core speech band of 300Hz-3kHz by 3-5dB to generate enhanced digital audio.

6. The method for enhancing the directivity and clarity of audio in a dual-channel hearing aid according to claim 5, characterized in that, The specific process of Bluetooth signal encapsulation is as follows: The connection status parameters fed back by the mobile phone are read in real time through the Type-C interface, and the frequency of lag is detected. times / minute or transmission rate At Mbps, the Bluetooth transmission protocol is switched from APT-X to SBC. If the input is enhanced digital audio in conference mode, the enhanced digital audio is encapsulated into an SBC format Bluetooth signal according to the SBC encoding format. If the input is enhanced digital audio in outdoor mode, it is encapsulated into an SBC format Bluetooth signal according to the same SBC encoding format. When the stuttering frequency... times / minute and transmission rate At Mbps, the APT-X protocol is maintained. If the input is enhanced digital audio in conference mode, the enhanced digital audio will be encapsulated into an APT-X format Bluetooth signal according to the APT-X encoding format. If the input is enhanced digital audio in outdoor mode, the enhanced digital audio will be encapsulated into an APT-X format Bluetooth signal according to the same APT-X encoding format.

7. A system for enhancing the directivity and clarity of audio in a hearing aid through dual channels, characterized in that, The system is used to perform a method for enhancing the directivity of dual-channel audio clarity in a hearing aid, as described in any one of claims 1-6.

8. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program that is executed by a processor to implement a method for enhancing the directivity of dual-channel audio clarity in a hearing aid as described in any one of claims 1-6.

Images