In-vehicle dynamic sound isolation method and system and vehicle

Abstract

The invention discloses an in-vehicle dynamic sound isolation method and system and a vehicle, and belongs to the technical field of in-vehicle sound processing, and the method comprises the steps: capturing sound signals of all parts in the vehicle; a deep learning model is adopted to classify the captured sound signals, and the output of the built-in sound system is adjusted according to the sound types of the sound signals; and adjusting the volume and the sound insulation degree of each sound control area based on the sound setting rule of each sound control area in the vehicle. According to the invention, external noise and non-driving-related sound interference in the vehicle can be effectively isolated, conversation privacy among passengers is protected by isolating and controlling sound transmission, various driving environments and changes of passenger demands can be automatically adapted, and an optimized sound management scheme is provided.

Description

Technical Field

[0001] The present invention relates to the technical field of in-vehicle sound processing, and more specifically, to an in-vehicle dynamic sound isolation method, system and vehicle. Background Art

[0002] In modern vehicle use, the in-car sound environment needs to adapt to the needs of different passengers. For example, when a passenger in the vehicle needs to rest or answer a call, the conversations of other passengers or the driver, the noise outside the car, etc. may interfere with him; or when passengers in the back seat talk about something involving privacy, it is necessary to avoid the driver in the front seat from hearing the relevant content as much as possible. Therefore, it is essential to equip the car with an intelligent sound isolation system that can dynamically adjust according to the needs of passengers to improve the comfort and driving safety in the car.

[0003] Existing solutions usually adopt the following two sound control technologies: (1) Passive sound insulation technology: using high-density materials such as sound insulation cotton, sound insulation felt, heavy curtains, double-layer sound insulation glass and other materials can effectively absorb or reflect sound waves and reduce the transmission of sound wave energy; (2) Active noise control technology: by emitting sound waves with opposite phase to the external noise, the noise can be reduced or eliminated. This technology mainly targets low-frequency noise.

[0004] However, passive sound insulation technology cannot adjust the sound insulation effect according to real-time needs and isolate all sounds equally; active noise control technology mainly targets low-frequency noise and cannot be adjusted according to the different needs of different passengers in the car. Summary of the invention

[0005] One object of the present invention is to provide a method, system and vehicle for dynamic sound isolation in a vehicle, so as to solve the problem that the aforementioned passive sound insulation technology cannot adjust the sound insulation effect according to real-time needs and isolate all sounds equally; the active noise control technology mainly targets low-frequency noise and cannot be adjusted according to the different needs of different passengers in the vehicle.

[0006] According to a first aspect of the present invention, there is provided a method for dynamic sound isolation in a vehicle, comprising:

[0007] Capture sound signals from all parts of the vehicle interior;

[0008] A deep learning model is used to classify the captured sound signals and adjust the output of the built-in sound system according to the sound type of the sound signal;

[0009] Based on the sound setting rules of each sound control area inside the vehicle, adjust the volume and sound insulation level of each sound control area.

[0010] Optionally, in the step of capturing sound signals at various locations inside the vehicle, multiple microphones are deployed at various locations inside the vehicle to obtain sound signals captured by each microphone.

[0011] Optionally, in the step of capturing sound signals at various locations inside the vehicle, signal comparison is performed on the sound signals captured by each microphone to obtain the time difference and intensity difference of the same sound signal, and the sound source position and sound source intensity are determined based on the time difference and intensity difference information.

[0012] Optionally, the step of using a deep learning model to classify the captured sound signal specifically includes:

[0013] Preprocess the captured sound signal, remove the background noise of the sound signal, enhance the specific characteristics of the sound signal, and divide the continuous sound signal into frames;

[0014] Short-time Fourier transform (STFT) is used to convert the sound signal from the time domain to the frequency domain to obtain the information of the frequency components;

[0015] The sequence data in the sound signal is processed by a recurrent neural network (RNN), the sound signal data is divided into a training set, a validation set and a test set, the training set data is used to train the RNN model, the model parameters are updated through the back propagation algorithm and the optimizer, the trained RNN model is evaluated using the test set, the evaluation index is calculated, the sound signal to be recognized is input into the trained RNN model, and the corresponding recognition result is output through the RNN model.

[0016] Optionally, the step of adjusting the output of the built-in sound system according to the sound type includes:

[0017] Analyze the frequency and phase of the sound signal and generate an anti-phase sound wave signal with opposite phase and similar amplitude to offset the intrusion noise; and / or,

[0018] The collected human voice signal in the passenger area is enhanced.

[0019] Optionally, the step of using a deep learning model to classify the captured sound signal and adjusting the output of the built-in sound system according to the sound type further includes:

[0020] Coding techniques based on psychoacoustic models identify specific frequencies, phases, and dynamic ranges in sound signals while removing redundant information;

[0021] An adaptive bitrate allocation mechanism is used to dynamically adjust the encoding bitrate of the sound signal to retain the key features of the audio signal.

[0022] Optionally, in the step of adjusting the volume and sound insulation level of each sound control area based on the sound setting rules of each sound control area inside the vehicle:

[0023] The sound setting rules for each sound control area are preset through the vehicle interface or mobile device. The sound setting rules include conditional triggers based on time, passenger activity and vehicle status. The volume, sound quality parameters and sound insulation effect of each sound area are automatically adjusted based on the activity status of passengers in the vehicle, the external environment and internal status of the vehicle and the sound setting rules.

[0024] Optionally, the in-vehicle dynamic sound isolation method further includes:

[0025] Receive user feedback and adjust preset sound setting rules, while learning user preferences to optimize subsequent sound settings.

[0026] According to a second aspect of the present invention, there is provided a vehicle interior dynamic sound isolation system, comprising:

[0027] A sound acquisition module, used to acquire sound signals from various locations inside the vehicle;

[0028] The sound analysis module uses a deep learning model to classify the captured sound signals and adjust the output of the built-in sound system according to the sound type;

[0029] The sound isolation module adjusts the volume and sound insulation level of each sound control area based on the sound setting rules of each sound control area inside the vehicle.

[0030] According to a third aspect of the present invention, there is provided a vehicle provided with the above-mentioned in-vehicle dynamic sound isolation system.

[0031] The in-vehicle dynamic sound isolation method, system and vehicle disclosed herein have the following technical effects:

[0032] A deep learning model is used to classify the captured sound signals and adjust the output of the built-in audio system according to the sound type. Based on the sound setting rules of each sound control area inside the vehicle, the volume and sound insulation level of each sound control area are adjusted.

[0033] The present invention can provide customized sound isolation according to the individual needs of different passengers. For example, when someone is on the phone or needs a quiet rest, the system should be able to automatically enhance the sound isolation effect of the passenger area without affecting other passengers.

[0034] The sound environment inside and outside the vehicle is constantly changing, and the present invention is able to monitor and quickly respond to these changes in real time, such as automatically enhancing the sound insulation effect when the vehicle enters a noisy environment, or reducing in-vehicle distractions when the driver needs to concentrate.

[0035] The present invention can identify and distinguish the sources and types of different sounds in the car (such as the driver's voice, passengers talking, babies crying, etc.) with high accuracy, and process these sounds appropriately, such as amplifying, weakening or eliminating specific sounds.

[0036] The present invention can create multiple independent sound zones in the car, allowing passengers in different zones to adjust the sound environment according to their respective needs, such as the front passengers listening to music while the back passengers watching movies, without being affected.

[0037] Further features and advantages of the present invention will become apparent from the following detailed description of exemplary embodiments of the present invention with reference to the attached drawings. BRIEF DESCRIPTION OF THE DRAWINGS

[0038] The accompanying drawings, which are incorporated in and constitute a part of the specification, illustrate embodiments of the invention and, together with the description, serve to explain the principles of the invention.

[0039] Figure 1 A schematic diagram of a flow chart of a method for dynamic sound isolation in a vehicle provided by an embodiment of the present invention;

[0040] Figure 2 A schematic diagram of the composition of an in-vehicle dynamic sound isolation system provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0041] Various exemplary embodiments of the present invention will now be described in detail with reference to the accompanying drawings. It should be noted that the relative arrangement of components and steps, numerical expressions and numerical values ​​set forth in these embodiments do not limit the scope of the present invention unless otherwise specifically stated.

[0042] The following description of at least one exemplary embodiment is merely illustrative in nature and is in no way intended to limit the invention, its application, or uses.

[0043] Technologies, methods, and equipment known to ordinary technicians in the relevant art may not be discussed in detail, but where appropriate, the technologies, methods, and equipment should be considered as part of the specification.

[0044] In all examples shown and discussed herein, any specific values ​​should be interpreted as merely exemplary and not limiting. Therefore, other examples of the exemplary embodiments may have different values.

[0045] In modern vehicle design, the optimization of the in-vehicle sound environment has become an important consideration for improving the riding experience and ensuring privacy protection. With the advancement of science and technology, especially the integration of acoustic technology and intelligent control systems, it is particularly important to develop an intelligent sound isolation system that can be dynamically adjusted according to the specific needs of passengers. This application strategically deploys multiple high-sensitivity microphones throughout the vehicle. Through the time difference and intensity difference of the same sound captured by different microphones, the position and intensity of the sound source can be accurately located. Combined with voiceprint recognition technology, the system can identify the voice of a specific user and provide support for personalized services. Use a well-trained deep learning model to classify the captured sound in real time, distinguish whether it is human voice, music, road noise, etc., and adopt different processing strategies accordingly.

[0046] The present invention proposes an embodiment of a method for dynamic sound isolation in a vehicle, specifically, as follows Figure 1 shown, including:

[0047] Step S1, capturing sound signals from various locations inside the vehicle;

[0048] It should be noted that this step relies on a high-sensitivity microphone array arranged at multiple locations inside the vehicle. These microphones can capture sound signals from different directions and frequencies, including but not limited to passenger conversations, road noise, engine noise, wind noise, music playing sound, etc.

[0049] Step S2, using a deep learning model to classify the captured sound signal, and adjusting the output of the built-in sound system according to the sound type of the sound signal;

[0050] It should be noted that a large amount of labeled sound signal data is used to train the deep learning model. The deep learning model can learn and identify different types of sound features, such as human voice, mechanical noise, environmental noise, etc. The deep learning model classifies the sound signals captured in real time. By comparing the characteristics of the sound signal with the learned sound template, the deep learning model can quickly and accurately determine the type and source of the current sound. Different adjustment strategies are formulated based on the classification results. For example, for human voice, the volume of certain areas can be selectively increased or decreased so that passengers can have better conversations or maintain privacy; for noise, corresponding sound insulation measures can be taken.

[0051] Step S3, based on the sound setting rules of each sound control area inside the vehicle, adjust the volume and sound insulation level of each sound control area.

[0052] It should be noted that the interior of the vehicle is divided into multiple sound control areas, such as the cockpit, passenger compartment, trunk, etc. A series of sound setting rules are set according to factors such as vehicle type, passenger preferences, and driving environment. The sound setting rules include but are not limited to: volume thresholds for different sound types in different areas, output optimization schemes for the audio system, etc. Based on the real-time sound classification results and the preset sound setting rules, the volume and sound insulation level of each sound control area are dynamically adjusted. For example, when strong engine noise is detected, the sound insulation effect of the cockpit and passenger compartment will be automatically increased; when passengers play music, the output effect of the audio system will be adjusted according to the type of music and passenger preferences.

[0053] In the embodiment of the present invention, in the step of capturing sound signals at various locations inside the vehicle, multiple microphones are deployed at various locations inside the vehicle to obtain the sound signals captured by each microphone. Multiple high-sensitivity microphones are strategically deployed at various locations inside the vehicle to ensure all-round sound capture, from passenger conversations to road noise. This multi-microphone array can capture sound signals from different directions, providing a rich data foundation for subsequent sound analysis and processing.

[0054] In an embodiment of the present invention, in the step of capturing sound signals at various locations inside the vehicle, a signal comparison is performed for the sound signals captured by each microphone, the time difference and intensity difference of the same sound signal are obtained, and the sound source position and sound source intensity are determined based on the time difference and intensity difference information, so as to identify the source of the sound and perform corresponding processing. Among them, the time difference refers to the time delay between the same sound signal reaching different microphones; the intensity difference refers to the energy difference received by the same sound signal at different microphones. By comparing the sound intensities received by different microphones, the results of the time difference analysis can be further verified and help to more accurately determine the location of the sound source. Based on the time difference and intensity difference information, a geometric positioning algorithm (such as hyperbolic positioning, spherical positioning, etc.) is used to determine the specific location of the sound source. This step is crucial for distinguishing internal sounds (such as the driver's commands or the conversations of passengers) from external noise (such as traffic noise). Once the location and intensity of the sound source are determined, subsequent processing can be performed as needed. For example, in automobile noise control, the output of the in-car audio system can be adjusted according to the location of the sound source or the active noise reduction system can be activated to reduce noise interference; in video conferencing, the camera's viewing angle can be adjusted according to the location of the sound source or the audio signal in a specific direction can be enhanced to improve the call quality.

[0055] In an embodiment of the present invention, the step of classifying the captured sound signal using a deep learning model specifically includes:

[0056] The captured sound signal is preprocessed to remove the background noise of the sound signal, enhance the specific characteristics of the sound signal, and segment the continuous sound signal into frames; wherein, the background noise of the sound signal is removed by a bandpass filter, a Wiener filter, or a spectral subtraction method, a subspace algorithm, etc.; the continuous sound signal is segmented into frames based on the short-time stationary characteristics of the sound signal, that is, in a short time (such as 10-30 milliseconds), the characteristics of the sound signal are relatively stable.

[0057] Short-time Fourier transform (STFT) is used to convert the sound signal from the time domain to the frequency domain to obtain information about the frequency components; the result of STFT is a complex matrix containing the amplitude and phase information of the sound signal.

[0058] The sequence data in the sound signal is processed by a recurrent neural network (RNN), and the sound signal data is divided into a training set, a validation set, and a test set. The training set data is used to train the RNN model, and the model parameters are updated through the back propagation algorithm and optimizer (such as Adam, SGD, etc.) to minimize the error. The trained RNN model is evaluated using the test set, and the evaluation indicators (such as accuracy, recall rate, F1 score, etc.) are calculated. The sound signal to be recognized is input into the trained RNN model, and the corresponding recognition result is output through the RNN model. Among them, the validation set is used to adjust the model parameters to prevent overfitting.

[0059] In an embodiment of the present invention, the step of adjusting the output of the built-in sound system according to the sound type includes:

[0060] The frequency and phase of the sound signal are analyzed to generate an anti-phase sound wave signal with an opposite phase and similar amplitude to the original noise signal to offset the invading noise. The anti-phase sound wave signal physically offsets the original noise signal, thereby reducing or eliminating the noise. The generated anti-phase sound wave signal is then played out by the speakers of the audio system and superimposed with the original noise signal in space to achieve active noise reduction.

[0061] And / or, the collected vocal signal in the passenger area is enhanced; wherein the gain is automatically adjusted according to the strength of the vocal signal to ensure that the clarity of the voice is maintained at different volumes, and the timbre of the vocal signal is optimized by adjusting the gain of different frequency bands to make it more pleasant to the ear.

[0062] In an embodiment of the present invention, the step of using a deep learning model to classify the captured sound signal and adjusting the output of the built-in sound system according to the sound type also includes:

[0063] Coding technology based on psychoacoustic models identifies specific frequencies, phases, and dynamic ranges in sound signals, while removing redundant information (such as noise, echo, etc.); among them, psychoacoustic models retain key phase information during the encoding process based on the impact of phase information on the overall perception of sound to maintain the spatial characteristics of sound; in areas with a larger dynamic range, coding technology based on psychoacoustic models uses finer quantization steps to retain more details, while in areas with a smaller dynamic range, coarser quantization is used to reduce bit consumption.

[0064] Adaptive bitrate allocation mechanism is used to dynamically adjust the encoding bitrate of the sound signal to preserve the key features of the audio signal; the adaptive bitrate allocation mechanism is an intelligent encoding strategy that dynamically adjusts the encoding bitrate according to the actual content and characteristics of the audio signal. In the adaptive bitrate allocation mechanism, key features refer to the factors that have the greatest impact on the perceived quality of the audio signal, such as the clarity of speech, the melody and rhythm of music, etc. By dynamically adjusting the bitrate, it can ensure that these key features are fully preserved during the compression process, so as to provide a high-quality audio experience while maintaining a lower bitrate.

[0065] In the embodiment of the present invention, in the step of adjusting the volume and the sound insulation level of each sound control area based on the sound setting rules of each sound control area inside the vehicle:

[0066] The sound setting rules for each sound control area are preset through the vehicle interface or mobile device. The sound setting rules include conditional triggers based on time, passenger activity and vehicle status; based on the activity status of passengers in the car, the external environment and internal status of the vehicle and the sound setting rules, the volume, sound quality parameters and sound insulation effect of each sound area are automatically adjusted. The coding technology based on the psychoacoustic model and the adaptive bit rate allocation mechanism achieve efficient compression and high-quality reconstruction of audio signals by simulating the auditory characteristics of the human ear and intelligently allocating bit resources.

[0067] In an embodiment of the present invention, the method for isolating dynamic sound in a vehicle further includes:

[0068] Receive user feedback and adjust the preset sound setting rules, and learn user preferences to optimize subsequent sound settings. Among them, the collected user feedback is recorded in detail and classified and sorted. During the analysis process, focus on the user's satisfaction with the sound isolation effect, the identification of specific noise sources (such as engine noise, wind noise, road noise, etc.), and suggestions for improvement of existing sound setting rules. Based on the long-term collected user feedback and data analysis results, the automobile manufacturer or audio system provider optimizes the preset sound setting rules for a long time, including but not limited to adjusting the sound isolation algorithm, optimizing the sound path design, improving the layout of the audio system, etc., to better adapt to the needs and preferences of different users. By analyzing the user's usage habits, sound setting adjustment records, and the user's response to specific sound types, the system gradually obtains the user's personalized needs and provides them with more personalized sound setting options. Based on the learned user preferences, the in-car audio system can intelligently recommend the most suitable sound setting solution for the user. For example, when the user selects a certain type of music, the equalizer setting will be automatically adjusted to highlight the characteristics of the music type; when the vehicle is driving under different road conditions, the sound isolation strategy will be adjusted according to the user's preferences to reduce external noise interference.

[0069] The embodiment of the present invention allows users to preset personalized sound preferences through the vehicle interface or mobile device, and can automatically adjust the sound processing strategy according to different driving scenarios (such as urban congested roads, highways, rural roads, parking and rest, etc.) to adapt to different noise environments and passenger needs. Among them, the vehicle's built-in intelligent system uses advanced machine learning algorithms to dynamically identify the current driving scene. Each driving scene has its unique noise characteristics and driving needs, and the system adjusts the sound processing strategy accordingly. When parking for a rest or traveling long distances, the user presets or automatically selects relaxing and pleasant environmental sound effects, such as natural scenery sounds, soft music, etc., to create a comfortable rest atmosphere and relieve driving fatigue. The in-vehicle dynamic sound isolation method of the present invention creates a more comfortable, safe and personalized in-vehicle sound environment for drivers and passengers by integrating personalized settings, intelligent scene recognition and automatic adjustment strategies, as well as continuous optimization and learning mechanisms.

[0070] The present invention also provides an embodiment of a dynamic sound isolation system in a vehicle, such as Figure 2 As shown, specifically including:

[0071] A sound acquisition module, used to acquire sound signals from various locations inside the vehicle;

[0072] The sound analysis module uses a deep learning model to classify the captured sound signals and adjust the output of the built-in sound system according to the sound type;

[0073] The sound isolation module adjusts the volume and sound insulation level of each sound control area based on the sound setting rules of each sound control area inside the vehicle.

[0074] It should be noted that the information interaction, execution process and other contents between the above-mentioned modules are based on the same concept as the method embodiment of the present application, and are a system corresponding to the above-mentioned in-vehicle dynamic sound isolation method. All implementation methods in the above-mentioned method embodiment are applicable to the embodiment of the device. Its specific functions and technical effects brought about can be found in the method embodiment part, which will not be repeated here.

[0075] The present invention also provides an embodiment of a vehicle, which is provided with the above-mentioned in-vehicle dynamic sound isolation system, wherein the vehicle is connected to a cloud server via an in-vehicle Internet, and regularly receives software updates and sound library expansion information, thereby ensuring the advancement and diversity of sound processing technology, and the user's driving experience will be continuously upgraded over time.

[0076] The in-vehicle dynamic sound isolation method, system and vehicle of the present invention can solve the problem of insufficient sound management in traditional vehicles and provide a more personalized and optimized sound environment. Each passenger is allowed to set the sound environment according to personal preferences to meet the sound needs of different passengers, thereby improving the riding experience. Dynamic sound management can effectively isolate external noise and non-driving-related sound interference in the car, helping the driver stay focused and reduce driving fatigue. By isolating and controlling the propagation of sound, the privacy of conversations between passengers is protected, especially in vehicles with multiple passengers. It can automatically adapt to changes in various driving environments and passenger needs, and can provide the best sound management solution whether it is fast driving on the highway or in urban congestion conditions.

[0077] The above describes in detail the structure, features and effects of the present invention based on the embodiments shown in the drawings, but the above is only a preferred embodiment of the present invention. It should be noted that the technical features involved in the above embodiments and their preferred methods can be reasonably combined and matched into a variety of equivalent schemes by those skilled in the art without departing from or changing the design ideas and technical effects of the present invention; therefore, the present invention is not limited to the scope of implementation shown in the drawings, and all changes made in accordance with the concept of the present invention, or modifications to equivalent embodiments with equivalent changes, which still do not exceed the spirit covered by the specification and drawings, should be within the protection scope of the present invention.

Claims

1. A method for isolating dynamic sound in a vehicle, characterized in that: include: Capture sound signals from all parts of the vehicle interior; A deep learning model is used to classify the captured sound signals and adjust the output of the built-in sound system according to the sound type of the sound signal; Based on the sound setting rules of each sound control area inside the vehicle, adjust the volume and sound insulation level of each sound control area.

2. The method for dynamic sound isolation in a vehicle according to claim 1, characterized in that: In the step of capturing sound signals at various locations inside the vehicle, multiple microphones are deployed at various locations inside the vehicle to obtain sound signals captured by each microphone.

3. The method for dynamic sound isolation in a vehicle according to claim 2, characterized in that: In the step of capturing sound signals at various locations inside the vehicle, signal comparison is performed on the sound signals captured by each microphone to obtain the time difference and intensity difference of the same sound signal, and the sound source position and sound source intensity are determined based on the time difference and intensity difference information.

4. The method for isolating dynamic sound in a vehicle according to claim 1, characterized in that: The steps of using a deep learning model to classify the captured sound signal include: Preprocess the captured sound signal, remove the background noise of the sound signal, enhance the specific characteristics of the sound signal, and divide the continuous sound signal into frames; Short-time Fourier transform (STFT) is used to convert the sound signal from the time domain to the frequency domain to obtain the information of the frequency components; The sequence data in the sound signal is processed by a recurrent neural network (RNN), the sound signal data is divided into a training set, a validation set and a test set, the training set data is used to train the RNN model, the model parameters are updated through the back propagation algorithm and the optimizer, the trained RNN model is evaluated using the test set, the evaluation index is calculated, the sound signal to be recognized is input into the trained RNN model, and the corresponding recognition result is output through the RNN model.

5. The method for dynamic sound isolation in a vehicle according to claim 4, characterized in that: In the steps to adjust the output of the built-in sound system according to the type of sound, include: Analyze the frequency and phase of the sound signal and generate an anti-phase sound wave signal with opposite phase and similar amplitude to offset the intrusion noise; and / or, The collected human voice signal in the passenger area is enhanced.

6. The method for isolating dynamic sound in a vehicle according to claim 5, characterized in that: The step of using a deep learning model to classify the captured sound signal and adjusting the output of the built-in sound system according to the sound type also includes: Coding techniques based on psychoacoustic models identify specific frequencies, phases, and dynamic ranges in sound signals while removing redundant information; An adaptive bitrate allocation mechanism is used to dynamically adjust the encoding bitrate of the sound signal to retain the key features of the audio signal.

7. The method for dynamic sound isolation in a vehicle according to claim 1, characterized in that: In the steps of adjusting the volume and sound insulation level of each sound control area based on the sound setting rules of each sound control area in the vehicle: The sound setting rules for each sound control area are preset through the vehicle interface or mobile device. The sound setting rules include conditional triggers based on time, passenger activity and vehicle status. The volume, sound quality parameters and sound insulation effect of each sound area are automatically adjusted based on the activity status of passengers in the vehicle, the external environment and internal status of the vehicle and the sound setting rules.

8. The method for isolating dynamic sound in a vehicle according to claim 7, characterized in that: Also includes: Receive user feedback and adjust preset sound setting rules, while learning user preferences to optimize subsequent sound settings.

9. A dynamic sound isolation system in a vehicle, characterized in that: include: A sound acquisition module, used to acquire sound signals from various locations inside the vehicle; The sound analysis module uses a deep learning model to classify the captured sound signals and adjust the output of the built-in sound system according to the sound type; The sound isolation module adjusts the volume and sound insulation level of each sound control area based on the sound setting rules of each sound control area inside the vehicle.

10. A vehicle, characterized in that: The in-vehicle dynamic sound isolation system according to claim 9 is provided.

Images