Audio adjusting method, electronic device, and storage medium

Abstract

The application discloses an audio adjusting method, an electronic device and a storage medium, and relates to the technical field of data processing. The method comprises the following steps: collecting a first audio signal representing environmental audio information, determining a first acoustic event corresponding to the first audio signal; determining an acoustic event sequence containing the first acoustic event and being sorted according to the occurrence time of the acoustic event; predicting the event change trend of each acoustic event in a preset time range after the current time according to the acoustic event sequence, and obtaining a prediction result, wherein the prediction result comprises the event type change trend, the event intensity change trend and the event occurrence probability of each acoustic event; determining an audio adjusting strategy corresponding to the prediction result, and adjusting the audio of a target device according to the audio adjusting strategy. The application improves the effect of audio adjustment for the target device.

Description

Technical Field

[0001] This application relates to the field of data processing technology, and in particular to audio adjustment methods, electronic devices, and storage media. Background Technology

[0002] Currently, for devices with audio capture and playback capabilities (such as headphones, especially active noise-canceling headphones), the approach is to switch between corresponding device modes based on environmental classification (such as subway, office, etc.), and adjust noise reduction based on the collected ambient noise intensity in different device modes (one type of audio adjustment). However, this approach has certain drawbacks. For example, it relies solely on simple environmental classification or the current real-time ambient noise intensity for audio adjustment, and cannot effectively adjust audio in complex mixed acoustic environments, thus resulting in the audio adjustment effect of the device not achieving the expected results.

[0003] Therefore, improving the effect of audio adjustment has become an urgent problem to be solved.

[0004] The above content is only used to help understand the technical solution of this application and does not represent an admission that the above content is prior art. Summary of the Invention

[0005] The main objective of this application is to provide an audio adjustment method, electronic device, and storage medium, aiming to solve the technical problem of how to improve the effect of audio adjustment.

[0006] To achieve the above objectives, this application proposes an audio adjustment method, which includes the following steps: Collect a first audio signal that characterizes environmental audio information, and determine the first acoustic event corresponding to the first audio signal; Determine the sequence of acoustic events ordered by their occurrence time, including the first acoustic event; Based on the acoustic event sequence, predict the event change trends of multiple acoustic events within a preset time range after the current time, and obtain the prediction results. The prediction results include the event type change trend, event intensity change trend, and event occurrence probability of each acoustic event. Determine the audio adjustment strategy corresponding to the prediction results, and adjust the audio of the target device according to the audio adjustment strategy.

[0007] Optionally, the step of determining an audio adjustment strategy corresponding to the prediction result and adjusting the audio of the target device according to the audio adjustment strategy includes: Determine the audio conditioning strategy corresponding to the prediction results, wherein the audio conditioning strategy includes a first strategy that characterizes the type of acoustic event that needs to be enhanced, a second strategy that characterizes the type of acoustic event that needs to be suppressed, and a third strategy that characterizes the allocation of priority weights to different acoustic events. The target device's audio is adjusted according to at least one of the first, second, and third strategies.

[0008] Optionally, the step of adjusting the audio of the target device according to the audio adjustment strategy includes: Determine the user preference model based on historical audio adjustment parameters for the target device; The audio adjustment strategy is input into the user preference model, and the updated audio adjustment strategy is output. The user preference model is used to update the adjustment amplitude and / or the priority weights of acoustic events according to the audio adjustment strategy. Perform the steps of adjusting the audio of the target device according to the updated audio adjustment strategy.

[0009] Optionally, the step of adjusting the audio of the target device according to the audio adjustment strategy further includes: Determine the audio adjustment parameters corresponding to the audio adjustment strategy, including noise reduction intensity parameters, transparency ratio parameters, and human voice enhancement coefficient; Adjust the audio of the target device based on the audio adjustment parameters.

[0010] Optionally, the step of adjusting the audio of the target device according to the audio adjustment parameters includes: Determine the rate of change of at least one acoustic event in an acoustic event sequence; When a change in the actual acoustic event corresponding to the target device is detected, the rate of change of the audio adjustment parameters in the target device is controlled according to the rate of change of the acoustic event.

[0011] Optionally, the step of adjusting the audio of the target device according to the audio adjustment parameters further includes: In response to the prediction results including confidence levels, it checks whether the confidence level is greater than a preset confidence threshold; If the confidence level is greater than the preset confidence threshold, the audio adjustment parameters are enhanced, and the audio of the target device is adjusted according to the enhanced audio adjustment parameters. If the confidence level is less than or equal to the preset confidence threshold, the audio of the target device will be adjusted according to the audio adjustment parameters.

[0012] Optionally, the step of predicting the event change trends of multiple acoustic events within a preset time range after the current time, based on the acoustic event sequence, and obtaining the prediction result, includes: Construct time series models corresponding to acoustic event sequences; Based on the time series model, predict the trend of event type change, event intensity change, and event occurrence probability of acoustic events within a preset time range after the current time, and use the predicted trend of event type change, event intensity change, and event occurrence probability of each acoustic event as the prediction result.

[0013] Optionally, the step of determining the first acoustic event corresponding to the first audio signal includes: Audio features are extracted from the first audio signal to obtain the first audio features, which include the spectral features, energy intensity features, temporal fluctuation features, and speech activity detection features of the first audio signal. Based on the preset mapping relationship between audio features and acoustic events, the first acoustic event corresponding to the first audio feature is determined. The event types of the acoustic events include acoustic events that characterize continuous background noise, acoustic events that characterize sudden transient sound, acoustic events that characterize human voice signals, and acoustic events that characterize wind noise signals.

[0014] In addition, to achieve the above objectives, this application also proposes an electronic device, which includes: a memory, a processor, and a computer program stored in the memory and executable on the processor, the computer program being configured to implement the steps of the audio adjustment method described above.

[0015] In addition, to achieve the above objectives, this application also proposes a storage medium, which is a computer-readable storage medium, on which a computer program is stored, and when the computer program is executed by a processor, it implements the steps of the audio adjustment method described above.

[0016] In this application, a first audio signal representing environmental audio information is acquired, and a first acoustic event corresponding to the first audio signal is identified. An acoustic event sequence containing the first acoustic event is then determined, ordered chronologically according to the occurrence time of the acoustic events. In other words, the first audio signal is directly associated with the first acoustic event, rather than categorizing the entire environment of the target device into a preset scene category. This is then further converted into a time-ordered sequence, i.e., an acoustic event sequence. Furthermore, when the acoustic environment is a complex mixed acoustic scene, multiple different types of acoustic events can be identified within the same audio signal, thereby achieving a multi-dimensional, fine-grained representation of the complex mixed acoustic scene. Based on this, a time sequence of these acoustic events can be constructed, facilitating subsequent audio adjustment strategies for the target device based on the complete acoustic event sequence rather than a single environment classification. In other words, by converting the audio signal into acoustic events for representation, the target device can achieve a refined description of complex mixed acoustic scenes, improving the accuracy of audio adjustment. It predicts the event change trends of multiple acoustic events within a preset time range after the current time based on the acoustic event sequence, obtaining prediction results including the event type change trend, event intensity change trend, and event occurrence probability of each acoustic event. Then, it adjusts the audio of the target device according to the corresponding audio adjustment strategy based on the prediction results. In other words, by predicting the trends of multiple acoustic events within a preset time range after the current time using an acoustic event sequence that includes the chronological order of the acoustic events, the audio adjustment strategy can be generated based on the upcoming acoustic environment. This realizes the transformation from delayed response to look-ahead pre-adjustment, eliminates the adjustment lag problem, and improves the audio adjustment effect for the target device. Attached Figure Description

[0017] The accompanying drawings, which are incorporated in and form part of this specification, illustrate embodiments consistent with this application and, together with the description, serve to explain the principles of this application.

[0018] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, for those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0019] Figure 1 This is a flowchart illustrating the first embodiment of the audio adjustment method of this application; Figure 2 This is a flowchart illustrating the second embodiment of the audio adjustment method of this application; Figure 3 This is a flowchart illustrating the third embodiment of the audio adjustment method of this application; Figure 4This is a flowchart illustrating the fourth embodiment of the audio adjustment method of this application; Figure 5 This is a schematic diagram of the device structure of the hardware operating environment involved in the audio adjustment method in the embodiments of this application.

[0020] The purpose, features, and advantages of this application will be further explained in conjunction with the embodiments and with reference to the accompanying drawings. Detailed Implementation

[0021] It should be understood that the specific embodiments described herein are merely illustrative of the technical solutions of this application and are not intended to limit this application.

[0022] To better understand the technical solution of this application, a detailed description will be provided below in conjunction with the accompanying drawings and specific implementation methods.

[0023] Current audio adjustment solutions for devices with audio capture and playback capabilities (such as headphones, especially active noise-canceling headphones) have the following shortcomings.

[0024] (1) The environmental modeling method is crude: it relies on the environmental classification results and cannot accurately describe complex mixed acoustic environments.

[0025] (2) Lack of time evolution capability: The audio is adjusted only based on the current environmental state and cannot predict future sound changes.

[0026] (3) Discretization of audio adjustment method: Currently, audio adjustment is achieved by mode switching, which cannot continuously and smoothly adjust the audio.

[0027] (4) Insufficient personalization capabilities: User behavior habits were not fully considered when adjusting audio.

[0028] Therefore, in order to avoid the above-mentioned defects, this application provides an audio adjustment method that can avoid the above-mentioned defects, and when the target device is headphones, the audio adjustment method can be a headphone audio adjustment method based on acoustic event sequence prediction.

[0029] Optionally, embodiments of this application may not rely on environmental classification, but instead collect environmental audio (such as a first audio signal), identify and construct an acoustic event sequence (a time series), predict acoustic change trends in the near future based on this acoustic event sequence, such as event change trends within a preset time range after the current time, and may combine a user preference model to dynamically generate priority-oriented audio adjustment strategies, and then continuously and smoothly adjust parameters such as noise reduction intensity, transparency ratio, and vocal enhancement of the target device (such as headphones) according to the audio adjustment strategies. In addition, embodiments of this application may also introduce adjustment transition mechanisms and confidence adjustment mechanisms to improve response speed and auditory experience.

[0030] Optionally, embodiments of this application can replace environmental classification and employ acoustic event sequence modeling (such as time series models) to improve the ability to represent complex environments. Furthermore, they can predict future changes in acoustic events based on acoustic event sequences, improving response foresight. Different priority weights can be assigned to different acoustic events to achieve differentiated processing. A preference model can be built based on user behavior, and the audio adjustment strategy can be adaptively optimized and adjusted according to the preference model, such as adjusting the audio adjustment parameters corresponding to the audio adjustment strategy. A transition mechanism can also be set to achieve smooth adjustment.

[0031] Based on this, the embodiments of this application provide an audio adjustment method, referring to... Figure 1 , Figure 1 This is a flowchart illustrating the first embodiment of the audio adjustment method of this application.

[0032] In this embodiment, the audio adjustment method includes steps S10 to S40.

[0033] Step S10: Acquire a first audio signal that characterizes environmental audio information and determine the first acoustic event corresponding to the first audio signal; Optionally, the first audio signal can be an ambient audio signal acquired in real time or periodically by an audio acquisition module (such as a microphone) built into or external to the target device (such as headphones). The first audio signal can represent all the sound information of the user's or target device's current environment, such as wind noise, human voices, machine noise, etc.

[0034] Optionally, the first acoustic event may be a single sound unit with semantic meaning identified from the first audio signal, and the first acoustic event may be stored in the form of text.

[0035] Optionally, the first acoustic event can be a type of acoustic event, which can be an event characterized by different sounds, such as human voice, wind noise, continuous background noise, and sudden transient sounds.

[0036] Optionally, each acoustic event may include the event type, event intensity, and occurrence time of the acoustic event.

[0037] Optionally, the event type can be an acoustic event type, such as an acoustic event representing a human voice signal corresponding to human voice, an acoustic event representing a wind noise signal corresponding to wind noise, an acoustic event representing a continuous background noise corresponding to continuous background noise, an acoustic event representing a sudden transient sound corresponding to sudden transient sound, etc.

[0038] Optionally, the event intensity can be the energy intensity of the acoustic event, such as 60 dB.

[0039] Optionally, the occurrence time may include the start time, end time, and duration of the acoustic event.

[0040] Alternatively, acoustic events can be represented in the form of text data in the target device.

[0041] Optionally, the audio adjustment method in this application embodiment can be applied to a target device, which may be an electronic device with audio acquisition and output, such as headphones or a mobile phone.

[0042] Alternatively, the audio adjustment method can be applied to an audio adjustment system.

[0043] Optionally, environmental audio can be acquired first to obtain a first audio signal that characterizes the environmental audio information.

[0044] Optionally, the first audio signal can be obtained by continuously acquiring ambient audio signals from outside the target device at a fixed sampling period (e.g., 50ms) through a built-in or external audio acquisition module (e.g., a microphone array) of the target device (e.g., headphones).

[0045] Optionally, acoustic event recognition and feature extraction can be performed on the acquired first audio signal to obtain the acoustic event corresponding to the first audio signal, and this event can be used as the first acoustic event.

[0046] Optionally, frequency analysis can be performed on the first audio signal, and the first acoustic event can be determined based on the frequency analysis results.

[0047] Optionally, an index corresponding to the first audio signal can be determined, and the index can be entered into a lookup table containing the mapping relationship between audio signals and acoustic events for searching. The acoustic event found can then be used as the first acoustic event.

[0048] Alternatively, other methods (such as neural networks) can be used to determine the first acoustic event corresponding to the first audio signal, without limitation.

[0049] Step S20: Determine the acoustic event sequence that is ordered according to the chronological order of the occurrence of the acoustic events and includes the first acoustic event; Optionally, the acoustic event sequence can be a mathematical structure formed by arranging multiple acoustic events identified within a duration window in chronological order of their occurrence.

[0050] Optionally, the acoustic event sequence can reflect the evolution of ambient sounds in the current environment over time.

[0051] For example, the acoustic events in the acoustic event sequence can be, in order: acoustic events representing persistent background noise, acoustic events representing sudden transient sound, acoustic events representing human voice signals, and acoustic events representing wind noise signals.

[0052] Optionally, an acoustic event sequence can be constructed, and the acoustic events identified within a continuous time window can be arranged in chronological order to form the sequence structure corresponding to Formula 1 below.

[0053] Formula 1; in, For acoustic event sequences, The earliest acoustic event in the sequence of acoustic events. This refers to the second acoustic event in the sequence of acoustic events. It is the latest acoustic event in the sequence of acoustic events, which can be the first acoustic event.

[0054] Optionally, the length of the acoustic event sequence can be determined based on a pre-set duration window, or it can be a pre-set fixed length, such as containing 5 acoustic events. Alternatively, different lengths of the acoustic event sequence can be set according to different environments. For example, in a relatively quiet scene, such as an indoor environment, the length of the acoustic event sequence can be set to the first length, while in a relatively noisy scene, such as a shopping mall environment, the length of the acoustic event sequence can be set to the second length, and the second length is greater than the first length.

[0055] Step S30: Based on the acoustic event sequence, predict the event change trends of multiple acoustic events within a preset time range after the current time, and obtain the prediction results; It should be noted that the prediction results include the trend of event type change, the trend of event intensity change, and the probability of event occurrence for each acoustic event; Optionally, the event change trend can be a prediction of the evolution of acoustic events within a preset time range after the current time. For example, it can be the event change trend of acoustic events within the next duration window, and the time interval of the next duration window can be the same as or match the preset time range. Optionally, the time length of the next duration window (i.e., the time interval, such as 30 seconds) can be the same as the time length of the duration window corresponding to the acoustic event sequence.

[0056] Optionally, the prediction results may include the event change trend of at least one acoustic event, specifically including the event type change trend, event intensity change trend, and event occurrence probability of at least one acoustic event.

[0057] Optionally, the trend of change in the event type of at least one acoustic event can be a prediction of an increase or decrease in the probability of the occurrence of at least one acoustic event (such as an acoustic event characterizing a human voice signal).

[0058] Optionally, the trend of change in the event intensity of at least one acoustic event can be a prediction of an increase or decrease in the sound pressure level or energy intensity of at least one acoustic event (such as an acoustic event characterizing wind noise signals).

[0059] Optionally, the probability of occurrence of at least one acoustic event can be the likelihood (e.g., 80%) of at least one acoustic event occurring within a preset time range after the current time.

[0060] Optionally, the preset time range can be a short window after the current time, or a long-duration window, such as 30 seconds. The size of the preset time range can be adjusted according to the user's needs.

[0061] Optionally, acoustic event trends can be predicted based on acoustic event sequences.

[0062] Optionally, acoustic event trend prediction can be performed on acoustic events of each event type in the acoustic event sequence to obtain prediction results.

[0063] Optionally, a machine learning model can be used for prediction. For example, an acoustic event sequence can be input into a machine learning model, and the output can be a prediction result representing the event change trend of multiple acoustic events within a preset time range after the current time.

[0064] Optionally, pre-set event change rules can be used to predict the event change trends of multiple acoustic events within a preset time range after the current time, based on the sorting order of each acoustic event in the acoustic event sequence.

[0065] Optionally, the event change rules can be pre-set fixed event change rules or event change rules determined based on the device parameters of the target device; there are no restrictions here. For example, the possible acoustic events in the prediction results can be acoustic events in the acoustic event sequence, or the occurrence time of each predicted acoustic event in the prediction results can be inversely proportional to the occurrence time of each acoustic event in the acoustic event sequence, etc.

[0066] Step S40: Determine the audio adjustment strategy corresponding to the prediction result, and adjust the audio of the target device according to the audio adjustment strategy.

[0067] Optionally, the audio conditioning strategy can be a set of control instructions generated based on the prediction results to guide the target device in adjusting audio output parameters. This audio conditioning strategy determines which sounds should be enhanced or suppressed, and the priority of the adjustments. For example, the audio conditioning strategy could assign corresponding priority weights to acoustic events.

[0068] For example, audio adjustment strategies could include maximizing active noise reduction on the target device in the presence of background noise.

[0069] Optionally, an audio adjustment strategy for the target device can be generated based on the prediction results.

[0070] For example, if the prediction results include continuous background noise events (i.e., acoustic events characterizing continuous background noise), the audio conditioning strategy can be active noise reduction processing by the target device; if the prediction results include acoustic events characterizing human voice signals, the audio conditioning strategy can be enabling transparency mode on the target device.

[0071] Optionally, an audio adjustment strategy for the target device can be generated based on at least one of the predicted acoustic event type change trend, event intensity change trend, and event occurrence probability.

[0072] Alternatively, different prediction results can generate different audio modulation strategies.

[0073] Optionally, the audio adjustment parameters (such as noise reduction intensity coefficient) of the target device can be determined based on the generated audio adjustment strategy (such as the audio adjustment strategy being to enhance human voice, suppress wind noise, and prioritize human voice). The audio adjustment of the target device can then be performed based on the audio adjustment parameters.

[0074] Optionally, the system can also determine the target device's current environmental information, such as whether it is in a subway, office, or outdoors, and adjust the audio adjustment strategy based on this information. Then, it can adjust the audio of the target device according to the adjusted audio adjustment strategy. For example, if the target device is in a subway, the audio adjustment strategy is strengthened; if the target device is in an office, the audio adjustment strategy is weakened.

[0075] In this embodiment, a first audio signal representing environmental audio information is acquired, and a first acoustic event corresponding to the first audio signal is determined. An acoustic event sequence containing the first acoustic event is then determined, ordered chronologically according to the occurrence time of the acoustic events. In other words, the first audio signal is directly associated with the first acoustic event, rather than classifying the entire environment of the target device into a preset scene category. This is then further converted into a time-ordered sequence, i.e., an acoustic event sequence. Furthermore, when the acoustic environment is a complex mixed acoustic scene, multiple different types of acoustic events can be identified in the same audio signal, thereby achieving a multi-dimensional, fine-grained representation of the complex mixed acoustic scene. Based on this, a time sequence of these acoustic events can be constructed, facilitating subsequent audio adjustment strategies for the target device based on the complete acoustic event sequence rather than a single environment classification. In other words, by converting the audio signal into acoustic events for representation, a refined description of the complex mixed acoustic scene can be achieved for the target device, improving the accuracy of audio adjustment. It predicts the event change trends of multiple acoustic events within a preset time range after the current time based on the acoustic event sequence, obtaining prediction results including the event type change trend, event intensity change trend, and event occurrence probability of each acoustic event. Then, it adjusts the audio of the target device according to the corresponding audio adjustment strategy based on the prediction results. In other words, by predicting the trends of each acoustic event within a preset time range after the current time using an acoustic event sequence that includes the chronological order of the acoustic events, the audio adjustment strategy can be generated based on the upcoming acoustic environment. This realizes the transformation from delayed response to look-ahead pre-adjustment, eliminates the adjustment lag problem, and improves the audio adjustment effect for the target device.

[0076] Based on the first embodiment of this application, a second embodiment of this application is proposed. In the second embodiment of this application, the same or similar contents as those in the above embodiment can be referred to the above description, and will not be repeated hereafter.

[0077] Reference Figure 2 In step S40, the step of determining the audio adjustment strategy corresponding to the prediction result and adjusting the audio of the target device according to the audio adjustment strategy includes steps a10 and a20.

[0078] Step a10: Determine the audio adjustment strategy corresponding to the prediction result; Step a20: Adjust the audio of the target device according to at least one of the first strategy, the second strategy, and the third strategy.

[0079] It should be noted that the audio conditioning strategy includes a first strategy that characterizes the types of acoustic events that need to be enhanced, a second strategy that characterizes the types of acoustic events that need to be suppressed, and a third strategy that characterizes the allocation of priority weights to different acoustic events. Optionally, the first strategy can be an enhancement strategy that characterizes the enhancement of acoustic events, allowing the target device to determine the level of perception of at least one acoustic event that needs to be improved. For example, when a human voice is predicted to appear, the transparency of the human voice signal should be enhanced so that the user can hear the conversation clearly.

[0080] Optionally, the second strategy can be a suppression strategy characterizing the suppression of acoustic events, which allows the target device to determine the perceived level of at least one acoustic event that needs to be reduced. For example, when a persistent background noise intensity is predicted, the noise reduction intensity should be increased to suppress the background noise.

[0081] Optionally, the third strategy can be a priority weighting strategy for representing acoustic events, which can assign different priority weights to different acoustic events. For example, the priority weight for acoustic events representing continuous background noise is 0.2, the priority weight for acoustic events representing sudden transient sounds is 0.1, the priority weight for acoustic events representing human voice signals is 0.6, and the priority weight for acoustic events representing wind noise signals is 0.1.

[0082] Optionally, after obtaining the prediction result based on the acoustic event sequence, a corresponding audio adjustment strategy can be generated based on the prediction result, and different prediction results correspond to different audio adjustment strategies.

[0083] Optionally, if the prediction results indicate that at least one acoustic event may occur within a preset time range after the current time, then at least one of the first, second, and third strategies can be used for audio adjustment of the target device. For example, if human voice signals (such as user voice), background noise, and wind noise signals may exist in the environment within a preset time range after the current time of the target device, then the first strategy can be used to enhance the human voice signal, the second strategy can be used to suppress the background noise, and a priority weight of 0.8 can be assigned to the human voice signal, a priority weight of 0.1 can be assigned to the background noise, and a priority weight of 0.1 can be assigned to the wind noise signal.

[0084] In this embodiment, by determining an audio adjustment strategy corresponding to the prediction result, such as a first strategy, a second strategy, and a third strategy, and by performing audio adjustment of the target device based on at least one of the first strategy, the audio adjustment of the target device can be refined based on different acoustic events, thereby improving the audio adjustment effect.

[0085] Optionally, in step S40, the step of adjusting the audio of the target device according to the audio adjustment strategy further includes steps b10-b30.

[0086] Step b10: Determine the user preference model based on the historical audio adjustment parameters for the target device; Step b20: Input the audio adjustment strategy into the user preference model and output the updated audio adjustment strategy; wherein, the user preference model is used to update the adjustment amplitude and / or the priority weight of acoustic events according to the audio adjustment strategy. Step b30: Perform the step of adjusting the audio of the target device according to the updated audio adjustment strategy.

[0087] Optionally, the historical audio adjustment parameters can be a record of the user's active adjustment of audio parameters during the use of the target device in a historical time range before the current time (such as the time range of the past week), such as noise reduction intensity, transparency ratio, and human voice enhancement coefficient.

[0088] Optionally, the user preference model can be a personalized model built based on users' historical behavior data, and can be a neural network, linear regression model, etc.

[0089] Optionally, the user preference model can learn users' preferences for different acoustic events.

[0090] Optionally, updating the adjustment magnitude can involve adjusting the execution strength of the audio adjustment strategy based on the user preference model. For example, if historical audio adjustment parameters indicate a user preference for aggressive adjustment, the magnitude of the enhancement and / or suppression strategies in the audio adjustment strategy can be increased. If historical audio adjustment parameters indicate a user preference for conservative adjustment, the magnitude of the enhancement and / or suppression strategies in the audio adjustment strategy can be decreased.

[0091] Optionally, the adjustment range can be adjusted to increase the volume and / or reduce noise of the target device. Optionally, the priority weights of acoustic events can be updated by adjusting the priority weights of each acoustic event according to user preferences. For example, if the user preference model learns from historical audio adjustment parameters that the user is particularly sensitive to acoustic events representing wind noise signals, the priority weight for suppressing wind noise signals can be increased.

[0092] Optionally, the system can record the user's audio adjustment behavior on the target device within a preset historical time range (such as the past week) (i.e., the user's historical audio adjustment behavior), and determine the historical audio adjustment parameters for the target device based on the user's historical audio adjustment behavior, with each user's historical audio adjustment behavior corresponding to a historical audio adjustment parameter.

[0093] Optionally, a user preference model can be constructed based on historical audio adjustment parameters. For example, multiple historical audio adjustment parameters can be used to train a pre-set model (such as a linear regression model, a lightweight neural network, etc.) to obtain a trained model, which can then be used as the user preference model.

[0094] Optionally, the audio adjustment strategy can be modified and updated using a user preference model to obtain an updated audio adjustment strategy.

[0095] For example, if there are 10 historical audio adjustment behaviors of a user, and the first 8 behaviors involved maximizing noise reduction in a windy environment, while the last 2 behaviors involved maintaining transparency in a human voice environment, a user preference model can be constructed based on the historical audio adjustment parameters corresponding to these 10 behaviors. This user preference model can emphasize wind noise sensitivity and moderate transparency for human voices. When the audio adjustment strategy is "human voice enhancement weight 0.7, wind noise suppression weight 0.5", the audio adjustment strategy can be input into the user preference model, and the updated audio adjustment strategy can be output as "human voice enhancement weight 0.6, wind noise suppression weight 0.9". The target device can then be audio adjusted according to this updated audio adjustment strategy.

[0096] Optionally, the human voice enhancement weight of 0.7 and the wind noise suppression weight of 0.5 can be used to enhance the acoustic events representing human voice signals according to the first strategy and assign a corresponding priority weight of 0.7, and to suppress the acoustic events representing wind noise signals according to the second strategy and assign a corresponding priority weight of 0.5.

[0097] Optionally, the human voice enhancement weight of 0.6 and the wind noise suppression weight of 0.9 can be used to enhance the acoustic events representing human voice signals according to the first strategy and assign a corresponding priority weight of 0.6, and to suppress the acoustic events representing wind noise signals according to the second strategy and assign a corresponding priority weight of 0.9.

[0098] In this embodiment, by updating the audio adjustment strategy based on the user preference model, and then adjusting the audio of the target device according to the updated audio adjustment strategy, the audio adjustment of the target device can conform to the user's personal habits, achieving adaptive audio adjustment and improving the personalization of audio adjustment.

[0099] Optionally, in step S40, the step of adjusting the audio of the target device according to the audio adjustment strategy further includes steps c10-c20.

[0100] Step c10: Determine the audio adjustment parameters corresponding to the audio adjustment strategy; Step c20: Adjust the audio of the target device according to the audio adjustment parameters.

[0101] It should be noted that the audio adjustment parameters include noise reduction intensity, transparency ratio, and vocal enhancement coefficient.

[0102] Optionally, the audio adjustment parameter can be an adjustable control quantity that directly acts on the target processing device. The audio adjustment parameter can be a continuously changing parameter.

[0103] Optionally, the noise reduction intensity parameter can be a continuous variable that controls the depth of active noise reduction.

[0104] Optionally, the transparency ratio parameter can be a continuous variable that controls the degree of ambient sound transmission.

[0105] Optionally, the voice enhancement factor can be a continuous variable that controls the gain of the voice frequency band.

[0106] Optionally, corresponding audio adjustment parameters can be determined based on the updated or adjusted audio adjustment strategy to adjust the audio of the target device.

[0107] Optionally, the noise reduction intensity parameter, transparency ratio parameter, and human voice enhancement coefficient of the target device can be adjusted, and different audio adjustment parameters can be adjusted for acoustic events in different time periods.

[0108] In this embodiment, the audio of the target device is adjusted according to the audio adjustment parameters corresponding to the audio adjustment strategy, thereby ensuring the effectiveness of the audio adjustment.

[0109] Optionally, step c20, which involves adjusting the audio of the target device based on the audio adjustment parameters, includes steps c21 and c22.

[0110] Step c21: Determine the rate of change of the acoustic event of at least one acoustic event in the acoustic event sequence; Step c22: When a change in the actual acoustic event corresponding to the target device is detected, the rate of change of the audio adjustment parameters in the target device is controlled according to the rate of change of the acoustic event.

[0111] Optionally, the rate of change of an acoustic event can be the speed at which the intensity or type of the acoustic event changes per unit time. For example, the rate of increase in intensity of a sudden transient sound (such as the sound of a door closing) can reach 100 dB / s, while the rate of change of background noise from the street into the room is about 5 dB / s.

[0112] Optionally, the rate of change can be the rate of change of the audio adjustment parameter over time.

[0113] Optionally, the fast response mode can allow audio adjustment parameters to change at a higher rate when the rate of change is greater than a first threshold (e.g., 30 dB / s) in order to quickly adapt to sudden environmental changes.

[0114] Optionally, the smooth transition mode can limit the rate of change of audio adjustment parameters when the rate of change is less than or equal to a first threshold, so as to achieve a seamless transition.

[0115] Optionally, transition control can be implemented to control the rate of change of audio adjustment parameters when the acoustic event changes. For example, if the acoustic event representing wind noise changes to the acoustic event representing human voice, the rate of change of the audio adjustment parameters corresponding to the acoustic event representing wind noise can be adjusted to the rate of change of the audio adjustment parameters corresponding to the acoustic event representing human voice.

[0116] Optionally, the actual acoustic event can be an acoustic event that has already occurred or is currently occurring, such as the first acoustic event.

[0117] Optionally, it is possible to statistically analyze the acoustic events of various event types that have occurred and determine the rate of change of the acoustic events per unit time to obtain the rate of change of the acoustic events.

[0118] Optionally, each acoustic event can correspond to an acoustic event change rate.

[0119] Optionally, the rate of change of audio adjustment parameters in the target device can be controlled according to the rate of change of at least one acoustic event. For example, the faster the rate of change of the acoustic event, the faster the rate of change of the audio adjustment parameters.

[0120] Optionally, the rate of change of the audio adjustment parameters can be adjusted according to the rate of change of the event, which can enable a rapid response to sudden acoustic events and a smooth transition to a gradually changing acoustic environment.

[0121] For example, if the background noise intensity changes from 50dB to 52dB within the last two duration windows, the rate of change is 2dB / s (below the threshold of 30dB / s). When the actual environment changes from quiet to street noise, the noise reduction intensity of the target device can be linearly increased from 0.2 to 0.8 at a rate of 0.05 / s, taking 12 seconds to achieve a smooth transition.

[0122] For example, upon detecting a sudden speech event (i.e., an acoustic event representing a human voice signal), the intensity jumps from 40dB to 90dB within 100ms, a rate of change of 500dB / s. The transparency ratio of the target device can be increased from 0.2 to 0.9 at a rate of 0.8 / s, with a response time of only 0.875 seconds, allowing the user to quickly hear the conversation.

[0123] In this embodiment, by determining the acoustic event change rate of at least one acoustic event, and controlling the change rate of audio adjustment parameters according to the acoustic event change rate when the acoustic event changes, the transition control of the acoustic event change scenario of the target device is realized, thereby improving the audio adjustment effect.

[0124] Optionally, step c20, which involves adjusting the audio of the target device based on the audio adjustment parameters, further includes steps c23 and c24.

[0125] Step c23: In response to the prediction result including confidence, check whether the confidence is greater than a preset confidence threshold; Step c24: If the confidence level is greater than the preset confidence threshold, then enhance the audio adjustment parameters and adjust the audio of the target device according to the enhanced audio adjustment parameters. Step c25: If the confidence level is less than or equal to the preset confidence threshold, then adjust the audio of the target device according to the audio adjustment parameters.

[0126] Optionally, if a model is used to predict the trend of at least one acoustic event, the output of the model (i.e. the prediction result) may include confidence levels.

[0127] Optionally, confidence level can be a characterization of the reliability of the prediction result.

[0128] Optionally, the preset confidence threshold can be the confidence threshold set in the extraction settings, such as 0.75.

[0129] Alternatively, conservative adjustment may involve either not enhancing the audio adjustment parameters or reducing them when the confidence level is less than or equal to a preset confidence threshold.

[0130] Optionally, when the confidence level of the prediction result exceeds a preset confidence threshold, audio adjustment parameters can be enhanced, such as increasing the voice enhancement coefficient or transparency ratio of the target device. Then, audio adjustment of the target device can be performed based on the enhanced audio adjustment parameters.

[0131] Optionally, when the confidence level of the prediction result is less than or equal to the preset confidence threshold, the audio adjustment parameters can be conservatively adjusted. For example, the audio adjustment parameters can be left unprocessed and the audio of the target device can be adjusted directly. Alternatively, the audio adjustment parameters can be suppressed, such as by weakening the audio adjustment parameters, and then the audio of the target device can be adjusted based on the weakened audio adjustment parameters.

[0132] Optionally, if the predicted acoustic event in the prediction result is substantially consistent with the corresponding acoustic event in the actual environment, then the confidence level is determined to be greater than the preset confidence threshold. For example, if it is predicted that speech will occur in 5 seconds, and the user's speech is detected in the actual environment 5 seconds later, then the confidence level is determined to be greater than the preset confidence threshold. Conversely, if the predicted event is not consistent with the actual event, then the confidence level is determined to be less than or equal to the preset confidence threshold.

[0133] In this embodiment, by including confidence in the prediction result, and enhancing the audio adjustment parameters when the confidence is greater than a preset confidence threshold, and adjusting the audio of the target device based on the enhanced audio adjustment parameters, or by directly adjusting the audio of the target device based on the audio adjustment parameters when the confidence is less than or equal to the preset confidence threshold, the audio adjustment effect for the target device can be improved.

[0134] Based on the first or second embodiment of this application, a third embodiment of this application is proposed. In the third embodiment of this application, the same or similar content as the above embodiment can be referred to the above description, and will not be repeated hereafter.

[0135] Reference Figure 3 In step S30, the step of predicting the event change trend of multiple acoustic events within a preset time range after the current time based on the acoustic event sequence, and obtaining the prediction result, includes steps d10-d20.

[0136] Step d10: Construct a time series model corresponding to the acoustic event sequence; Step d20: Based on the time series model, predict the event type change trend, event intensity change trend, and event occurrence probability of acoustic events within a preset time range after the current time, and use the predicted event type change trend, event intensity change trend, and event occurrence probability of each acoustic event as the prediction result.

[0137] Optionally, the time series model can be a model used to analyze and predict acoustic event sequences ordered by time, and can be a statistical model, a machine learning model, a recurrent neural network model, a long short-term memory network model, etc.

[0138] Optionally, the input to the time series model can be a sequence of acoustic events, and the output can be a prediction result.

[0139] Optionally, the trend of event type change can be predicted using a classification method, such as a time series model, to output the event type of the most likely acoustic event within a preset time range after the current time, or the direction of change in the probability of occurrence of each acoustic event. For example, the trend of change from acoustic events representing persistent background noise to acoustic events representing human voice signals.

[0140] Optionally, the trend of event intensity change can be predicted using a classification method, such as a time series model, which outputs the intensity estimates of each acoustic event within a preset time range after the current time, and determines the trend of event intensity change, such as rising or falling, based on the intensity value of at least one acoustic event in the acoustic event sequence and the intensity estimate belonging to that at least one acoustic event.

[0141] Optionally, the probability of an event can be predicted using a classification method, such as a time series model, to output the probability of each acoustic event occurring within a preset time range after the current time. For example, the probability of an acoustic event representing wind noise is 0.3.

[0142] Alternatively, the acoustic event sequence can be converted into a fixed-length feature vector sequence, in which each time step contains a one-hot encoding of the event type of the acoustic event and a normalized intensity value.

[0143] Optionally, a machine learning model can be constructed and trained to obtain a time series model. The feature vector sequence transformed from the acoustic event sequence is input into the time series model for prediction, so as to obtain the event type change trend, event intensity change trend and event occurrence probability of at least one acoustic event within a preset time range after the current time, and use them as the prediction result.

[0144] In this embodiment, by constructing a time series model corresponding to the acoustic time series, and then making predictions based on the time series model, the event type change trend, event intensity change trend, and event occurrence probability of the acoustic event are obtained, and these are used as the prediction results, thereby ensuring the accuracy of the obtained prediction results.

[0145] Based on any of the embodiments in the first to third embodiments of this application, a fourth embodiment of this application is proposed. In the fourth embodiment of this application, the same or similar content as the above embodiments can be referred to the above description, and will not be repeated hereafter.

[0146] Reference Figure 4 In step S10, the step of determining the first acoustic event corresponding to the first audio signal includes steps e10-e20.

[0147] Step e10: Extract audio features from the first audio signal to obtain the first audio features; Step e20: Based on the preset mapping relationship between audio features and acoustic events, determine the first acoustic event corresponding to the first audio feature.

[0148] It should be noted that the first audio features include the spectral features, energy intensity features, temporal fluctuation features, and speech activity detection features of the first audio signal.

[0149] Optionally, the event types of acoustic events include acoustic events characterizing persistent background noise, acoustic events characterizing sudden transient sound, acoustic events characterizing human voice signals, and acoustic events characterizing wind noise signals.

[0150] Optionally, the spectral characteristics can be frequency distribution information extracted by performing a short-time Fourier transform or Mel frequency cepstral coefficient processing on the first audio signal. Different timbres can be distinguished based on the spectral characteristics, such as differentiating human voice signals from wind noise signals in the first audio signal based on different spectral characteristics.

[0151] Optionally, the energy intensity characteristic can be the short-time energy or root mean square of the audio signal, reflecting the loudness of the sound.

[0152] Optionally, the time-domain fluctuation characteristics can be the rate of change of signal amplitude, zero-crossing rate, etc., used to identify sudden transient sounds or continuous background noise in the first audio signal.

[0153] Optionally, the speech activity detection feature can be the probability value of the signal containing speech calculated by performing speech activity detection on the first audio signal, that is, the probability value of determining that the first audio signal contains a human voice signal (that is, a signal representing speech).

[0154] Optionally, the mapping relationship can be a pre-defined mapping relationship between audio features and acoustic events, and can be stored in a table. Different audio features can correspond to different acoustic events.

[0155] Optionally, the first audio signal within the acquired duration window can be subjected to Fourier transform, frequency domain analysis can be performed, and corresponding audio features such as spectral features, energy intensity features, temporal fluctuation features, and speech activity detection features can be extracted and used as the first audio feature.

[0156] For example, the STFF can be calculated on the first audio signal to obtain spectral characteristics, the RMS can be calculated on the first audio signal to obtain energy intensity, the zero-crossing rate can be calculated on the first audio signal to obtain time-domain fluctuations, and a lightweight VAD model can be run to obtain speech probability.

[0157] Optionally, the first acoustic event can be determined based on a decision tree classifier that represents a preset mapping relationship between audio features and acoustic events. For example, the first audio feature can be input into the decision tree classifier, and the output can be the first acoustic event corresponding to the first audio feature. Alternatively, the first acoustic event corresponding to the first audio feature can be obtained by searching in a table containing preset mapping relationships between audio features and acoustic events based on the first audio feature.

[0158] In this embodiment, the first audio feature is obtained by extracting audio features from the first audio signal, and the first acoustic event is determined based on the mapping relationship between the audio feature and the acoustic event, thereby ensuring the accuracy of the determined first acoustic event.

[0159] In addition, this application also provides an audio adjustment system, which may include: The audio acquisition module is used to acquire a first audio signal that represents environmental audio information and to determine a first acoustic event corresponding to the first audio signal. The determination module is used to determine the sequence of acoustic events ordered according to the chronological order of their occurrence, and including the first acoustic event. The prediction module is used to predict the event change trend of at least one acoustic event within a preset time range after the current time based on the acoustic event sequence, and to obtain the prediction result, wherein the prediction result includes the event type change trend, event intensity change trend and event occurrence probability of at least one acoustic event; The audio adjustment module is used to determine the audio adjustment strategy corresponding to the prediction result, and to adjust the audio of the target device according to the audio adjustment strategy.

[0160] The audio adjustment system provided in this application, employing the audio adjustment method described in the above embodiments, can improve the audio adjustment effect. Compared with the prior art, the beneficial effects of the audio adjustment system provided in this application are the same as those of the audio adjustment method provided in the above embodiments, and other technical features of the audio adjustment system are the same as those disclosed in the methods of the above embodiments, and will not be repeated here.

[0161] Furthermore, this application provides an electronic device, which includes: at least one processor; and a memory communicatively connected to the at least one processor; wherein the memory stores instructions executable by the at least one processor, the instructions being executed by the at least one processor to enable the at least one processor to perform the audio adjustment method in the first embodiment described above.

[0162] The following is for reference. Figure 5The figure illustrates a structural diagram of an electronic device suitable for implementing embodiments of this application. The electronic devices in the embodiments of this application may include, but are not limited to, mobile terminals such as mobile phones, laptops, digital broadcast receivers, PDAs (Personal Digital Assistants), PADs (Portable Application Description), PMPs (Portable Media Players), in-vehicle terminals (e.g., in-vehicle navigation terminals), and fixed terminals such as digital TVs and desktop computers. The devices shown in the figure are merely examples and should not impose any limitations on the functionality and scope of use of the embodiments of this application.

[0163] The electronic device may include a processing unit 1001 (e.g., a central processing unit, a graphics processing unit, etc.) that can perform various appropriate actions and processes according to a program stored in a read-only memory (ROM) 1002 or a program loaded from a storage device 1003 into a random access memory (RAM) 1004. The RAM 1004 also stores various programs and data required for device operation. The processing unit 1001, ROM 1002, and RAM 1004 are interconnected via a bus 1005. An input / output (I / O) interface 1006 is also connected to the bus. Typically, the following systems can be connected to the I / O interface 1006: input devices 1007 including, for example, touchscreens, touchpads, keyboards, mice, image sensors, microphones, accelerometers, gyroscopes, etc.; output devices 1008 including, for example, liquid crystal displays (LCDs), speakers, vibrators, etc.; storage devices 1003 including, for example, magnetic tapes, hard disks, etc.; and communication devices 1009. Communication device 1009 allows electronic devices to communicate wirelessly or wiredly with other devices to exchange data. While electronic devices with various systems are shown in the figures, it should be understood that implementation or possession of all the systems shown is not required. More or fewer systems may be implemented alternatively.

[0164] Specifically, according to the embodiments disclosed in this application, the processes described above with reference to the flowcharts can be implemented as computer software programs. For example, embodiments disclosed in this application 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 device, or installed from storage device 1003, or installed from ROM 1002. When the computer program is executed by processing device 1001, it performs the functions defined in the methods of the embodiments disclosed in this application.

[0165] The electronic device provided in this application, employing the audio adjustment method in the above embodiments, can solve the technical problem of how to improve the effect of audio adjustment. Compared with the prior art, the beneficial effects of the electronic device provided in this application are the same as those of the audio adjustment method provided in the above embodiments, and other technical features of the electronic device are the same as those disclosed in the method of the previous embodiment, and will not be repeated here.

[0166] It should be understood that the various parts disclosed in this application can be implemented using hardware, software, firmware, or a combination thereof. In the description of the above embodiments, specific features, structures, materials, or characteristics can be combined in any suitable manner in one or more embodiments or examples.

[0167] The above are merely specific embodiments of this application, but the scope of protection of this application is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in this application should be included within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.

[0168] This application provides a computer-readable storage medium having computer-readable program instructions (i.e., a computer program) stored thereon, the computer-readable program instructions being used to perform the audio adjustment method in the above embodiments.

[0169] The computer-readable storage medium provided in this application may be, for example, a USB flash drive, but is not limited to, electrical, magnetic, optical, electromagnetic, infrared, or semiconductor systems, devices, 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 wires, 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 embodiment, the computer-readable storage medium may be any tangible medium containing or storing a program that can be used by or in conjunction with an instruction execution system, system, or device. The program code contained on the computer-readable storage medium may be transmitted using any suitable medium, including but not limited to: wires, optical cables, RF (Radio Frequency), etc., or any suitable combination thereof.

[0170] The aforementioned computer-readable storage medium may be included in an electronic device or may exist independently without being assembled into an electronic device.

[0171] The aforementioned computer-readable storage medium carries one or more programs, which, when executed by an electronic device, enable the electronic device to perform the steps in the aforementioned audio adjustment method.

[0172] Computer program code for performing the operations of this application can be written in one or more programming languages ​​or a combination thereof, including object-oriented programming languages ​​such as Java, Smalltalk, and C++, and conventional procedural programming languages ​​such as the "C" language or similar programming languages. The program code can be executed entirely on the user's computer, partially on the user's computer, as a standalone software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In cases involving remote computers, the remote computer can be connected to the user's computer via any type of network—including a Local Area Network (LAN) or a Wide Area Network (WAN)—or can be connected to an external computer (e.g., via the Internet using an Internet service provider).

[0173] 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 this application. 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.

[0174] The modules described in the embodiments of this application can be implemented in software or hardware. The names of the modules do not necessarily limit the functionality of the unit itself.

[0175] The readable storage medium provided in this application is a computer-readable storage medium that stores computer-readable program instructions (i.e., a computer program) for performing the above-described audio adjustment method, thereby solving the technical problem of how to improve the effect of audio adjustment. Compared with the prior art, the beneficial effects of the computer-readable storage medium provided in this application are the same as the beneficial effects of the audio adjustment method provided in the above embodiments, and will not be repeated here.

[0176] This application also provides a computer program product, including a computer program that, when executed by a processor, implements the steps of the audio adjustment method described above.

[0177] The computer program product provided in this application can solve the technical problem of how to improve the effect of audio adjustment. Compared with the prior art, the beneficial effects of the computer program product provided in this application are the same as the beneficial effects of the audio adjustment method provided in the above embodiments, and will not be repeated here.

[0178] The above are only some embodiments of this application and do not limit the patent scope of this application. All equivalent structural transformations made under the technical concept of this application and using the contents of the specification and drawings of this application, or direct / indirect applications in other related technical fields, are included in the patent protection scope of this application.

Claims

1. An audio adjustment method, characterized in that, The audio adjustment method includes the following steps: Collect a first audio signal that characterizes environmental audio information, and determine a first acoustic event corresponding to the first audio signal; Determine the sequence of acoustic events ordered by their occurrence time, including the first acoustic event; Based on the acoustic event sequence, the event change trends of multiple acoustic events within a preset time range after the current time are predicted to obtain prediction results. The prediction results include the event type change trend, event intensity change trend, and event occurrence probability of each acoustic event. The multiple acoustic events include acoustic events representing persistent background noise, acoustic events representing sudden transient sounds, acoustic events representing human voice signals, and acoustic events representing wind noise signals. Different environments correspond to different lengths of acoustic event sequences, and the second length of the acoustic event sequence in a shopping mall environment is greater than the first length of the acoustic event sequence in an indoor environment. Determine an audio adjustment strategy corresponding to the prediction result, wherein the audio adjustment strategy includes a first strategy characterizing the type of acoustic event that needs to be enhanced, a second strategy characterizing the type of acoustic event that needs to be suppressed, and a third strategy characterizing the allocation of priority weights to different acoustic events. Determine the audio adjustment parameters corresponding to the audio adjustment strategy, wherein the audio adjustment parameters include noise reduction intensity parameters, transparency ratio parameters, and human voice enhancement coefficient; Determine the rate of change of at least one acoustic event among the acoustic events; When a change in the actual acoustic event corresponding to the target device is detected, the rate of change of the audio adjustment parameters in the target device for audio adjustment is controlled according to the rate of change of the acoustic event. In response to the prediction result including a confidence level, it is detected whether the confidence level is greater than a preset confidence threshold; If the confidence level is greater than a preset confidence threshold, the audio adjustment parameters are enhanced, and the audio of the target device is adjusted according to the enhanced audio adjustment parameters. If the confidence level is less than or equal to the preset confidence threshold, then the audio of the target device is adjusted according to the audio adjustment parameters.

2. The audio adjustment method as described in claim 1, characterized in that, The step of adjusting the audio of the target device according to the audio adjustment strategy includes: Determine the user preference model based on historical audio adjustment parameters for the target device; The audio adjustment strategy is input into the user preference model, and an updated audio adjustment strategy is output; wherein, the user preference model is used to update the adjustment amplitude and / or the priority weight of acoustic events according to the audio adjustment strategy; The step of adjusting the audio of the target device according to the updated audio adjustment strategy is performed.

3. The audio adjustment method as described in claim 1, characterized in that, The step of predicting the event change trends of multiple acoustic events within a preset time range after the current time based on the acoustic event sequence, and obtaining the prediction result, includes: Construct a time series model corresponding to the acoustic event sequence; Based on the time series model, the event type change trend, event intensity change trend, and event occurrence probability of the acoustic events within a preset time range after the current time are predicted, and the predicted event type change trend, event intensity change trend, and event occurrence probability of each acoustic event are used as the prediction results.

4. The audio adjustment method as described in claim 1, characterized in that, The step of determining the first acoustic event corresponding to the first audio signal includes: Audio features are extracted from the first audio signal to obtain first audio features, wherein the first audio features include the spectral features, energy intensity features, temporal fluctuation features, and speech activity detection features of the first audio signal; Based on the preset mapping relationship between audio features and acoustic events, a first acoustic event corresponding to the first audio feature is determined. The event types of the acoustic events include acoustic events representing continuous background noise, acoustic events representing sudden transient sounds, acoustic events representing human voice signals, and acoustic events representing wind noise signals.

5. An electronic device, characterized in that, The electronic device includes: a memory, a processor, and a computer program stored in the memory and executable on the processor, the computer program being configured to implement the steps of the audio adjustment method as described in any one of claims 1 to 4.

6. A storage medium, characterized in that, The storage medium is a computer-readable storage medium, and a computer program is stored on the storage medium. When the computer program is executed by a processor, it implements the steps of the audio adjustment method as described in any one of claims 1 to 4.

Images