Broadcast volume adaptive adjustment method and device, equipment, medium and product

Abstract

The invention discloses a broadcast volume adaptive adjustment method and device, equipment, a medium and a product. Comprising the following steps: acquiring a partition sound signal, and preprocessing the acquired sound signal to acquire partition sound data; performing feature extraction on the acquired sound data, separating environmental noise and broadcast signals based on the extracted comprehensive features, and respectively measuring corresponding average sound pressure levels for the separated environmental noise and broadcast signals; determining a corresponding optimal parameter strategy based on the acquired average sound pressure level, the broadcast content type and the acquired environment data in combination with a reinforcement learning algorithm; and adjusting broadcast parameters based on the optimal parameter strategy. The adaptive adjustment of the broadcast volume based on the optimal parameter strategy is realized, the auditory experience of audiences in a complex acoustic environment is ensured, the optimal propagation effect of broadcast signals under different conditions is ensured, and the acquisition of key information by the audiences is ensured.

Description

Technical Field

[0001] The present invention relates to the technical field of broadcasting equipment, and in particular to a method, device, equipment, medium and product for adaptively adjusting broadcasting volume. Background Art

[0002] Nowadays, broadcasting systems are widely used in various fields. In the actual application of broadcasting systems, whether it is open spaces such as outdoor squares, stations, parks, or public areas such as schools, airports, and subways, broadcasting systems are faced with complex acoustic environments. These areas are often accompanied by huge traffic flows, and the noise generated cannot be ignored. When the noise is too loud, the broadcast sound is often easily covered, resulting in the inability of the audience to accurately receive and understand the broadcast information.

[0003] At present, when the ambient noise is too loud, the broadcast volume is adjusted to cover the noise. In the existing technology, this is often done through manual adjustment or automatic system adjustment. The manual adjustment method requires a lot of manpower output. For the existing automatic adjustment, it is often difficult to separate the noise and the broadcast. The adjustment process involves fewer broadcast parameters, and it is impossible to perform adaptive adjustments in real time as the noise level changes. Moreover, the adjustment effect is difficult to achieve a satisfactory auditory experience for the audience. Summary of the invention

[0004] The present invention provides a method, device, equipment, medium and product for adaptively adjusting broadcast volume, so as to solve the problems in the prior art that noise and broadcast separation technology are inaccurate and the adaptive adjustment effect of broadcast parameters cannot meet user needs.

[0005] In a first aspect, an embodiment of the present invention provides a method for adaptively adjusting broadcast volume, comprising:

[0006] Acquire partition sound signals, and pre-process the acquired partition sound signals to acquire partition sound data;

[0007] Extracting features from the acquired sound data to obtain comprehensive features, separating the ambient noise and the broadcast signal based on the acquired comprehensive features, and measuring the corresponding average sound pressure levels of the separated ambient noise and the broadcast signal respectively;

[0008] Adaptively adjust the optimal parameter strategy corresponding to the broadcast based on the acquired environmental noise and average sound pressure level of the broadcast signal, the type of broadcast content, the collected environmental data and the reinforcement learning algorithm;

[0009] Based on the optimal parameter strategy, current parameters of the broadcast signal are adjusted to parameters corresponding to the optimal parameter strategy.

[0010] As a preferred embodiment of the first aspect, the comprehensive features include spectrum features, time domain features and sound signal energy features.

[0011] As a preferred embodiment of the first aspect, the separation of ambient noise and broadcast signal based on the acquired comprehensive features specifically includes the following steps:

[0012] When the broadcasting system is not working, the subarea sound data is the first environmental noise of the corresponding subarea;

[0013] When the broadcasting system is working, the partition sound data is obtained in real time, and the extracted comprehensive features are input into a preset sound classification model to obtain classification information of the partition sound data;

[0014] For the part classified and marked as noise, the residual broadcast signal components are removed by filtering;

[0015] For the part classified and marked as broadcast signal, the residual noise component is removed by filtering;

[0016] The partitioned sound data and the corresponding classification labeling results are input into a separation algorithm, and the separated second environmental noise and the first broadcast signal are output.

[0017] As a preferred embodiment of the first aspect, the measuring corresponding average sound pressure level specifically includes: dividing the second environmental noise and the first broadcast signal into time windows, and calculating the average sound pressure level corresponding to all time serial ports.

[0018] As a preferred embodiment of the first aspect, the adaptively adjusting the optimal parameter strategy corresponding to the broadcast based on the acquired environmental noise and average sound pressure level of the broadcast signal, the broadcast content type, the collected environmental data and the reinforcement learning algorithm includes:

[0019] Defining the environmental state of the reinforcement learning algorithm based on the current environmental noise and the average sound pressure level of the broadcast signal, the type of broadcast content, and the collected environmental data; wherein the collected environmental data includes the flow of people in the partition and the time;

[0020] Defining an action space of the reinforcement learning algorithm according to broadcast parameters; wherein the action space includes: adjusting the volume, gain ratio and corresponding frequency response of the broadcast signal;

[0021] Determine a reward value corresponding to a specific execution action of a current broadcast signal parameter based on the environment state, the action space, and a reward function in the reinforcement learning algorithm;

[0022] Based on the reward value, the current parameter strategy of the broadcast signal is optimized to obtain the optimal parameter strategy.

[0023] As an optional embodiment of the first aspect, adjusting the current parameter of the broadcast signal to a parameter corresponding to the optimal parameter strategy based on the optimal parameter strategy includes:

[0024] Sending the optimal parameter strategy to the broadcast to be adjusted, and determining the volume, gain ratio and frequency response corresponding to the broadcast to be adjusted in the optimal parameter strategy;

[0025] Based on the dynamic adjustment algorithm, the current parameters of the broadcast to be adjusted are adjusted to the corresponding parameters in the optimal parameter strategy.

[0026] In a second aspect, an embodiment of the present invention provides a broadcast volume adaptive adjustment device, the broadcast volume adaptive adjustment device comprising a real-time acquisition module, a separation measurement module, a strategy determination module and a broadcast adjustment module;

[0027] The real-time acquisition module is used to obtain partition sound signals and pre-process the obtained partition sound signals to obtain partition sound data;

[0028] The separation and measurement module extracts features from the acquired sound data to obtain comprehensive features, separates the ambient noise and the broadcast signal based on the acquired comprehensive features, and measures the corresponding average sound pressure levels of the separated ambient noise and the broadcast signal respectively;

[0029] The strategy determination module is used to adaptively adjust the optimal parameter strategy corresponding to the broadcast according to the acquired environmental noise and average sound pressure level of the broadcast signal, the type of broadcast content, the collected environmental data and the reinforcement learning algorithm;

[0030] The broadcast adjustment module is used to adjust the current parameters of the broadcast signal to the parameters corresponding to the optimal parameter strategy according to the optimal parameter strategy.

[0031] In a third aspect, an embodiment of the present invention further provides an electronic device, the electronic device comprising:

[0032] one or more processors;

[0033] A storage medium for storing one or more programs;

[0034] When the one or more programs are executed by the one or more processors, the one or more processors implement the broadcast volume adaptive adjustment method described in any embodiment of the present invention.

[0035] In a fourth aspect, an embodiment of the present invention further provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the broadcast volume adaptive adjustment method described in any embodiment of the present invention.

[0036] In a fifth aspect, an embodiment of the present invention further provides a computer program product, including a computer program, and when the computer program is executed by a processor, the method for adaptively adjusting the broadcast volume as described in any embodiment of the present invention is implemented.

[0037] The embodiments of the present invention provide a method, device, electronic device, storage medium and product for adaptively adjusting broadcast volume. The method obtains sound signals in different partitions, pre-processes the acquired partition sound signals to obtain partition sound data, extracts features from the acquired sound data to obtain comprehensive features, and can more effectively separate noise and broadcast based on the comprehensive features. The corresponding average sound pressure levels of the separated environmental noise and broadcast signals are respectively measured, and the sound pressure level data is used as a basis for adjusting broadcast parameters and environmental control to ensure the quality of sound propagation and help evaluate the propagation effect of broadcast signals. The corresponding optimal parameter strategy is determined by a reinforcement learning algorithm based on the acquired environmental noise in the partition and the average sound pressure level of the broadcast signal, the type of broadcast content and other data, and the broadcast parameters are adaptively and dynamically adjusted based on the dynamic adjustment technology to ensure that the audience's auditory experience is met in a complex acoustic environment, ensure the best propagation effect of the broadcast signal under different circumstances, and ensure the audience's acquisition of key information. BRIEF DESCRIPTION OF THE DRAWINGS

[0038] In order to more clearly illustrate the technical solution of the present application, the drawings required for use in the implementation manner will be briefly introduced below. Obviously, the drawings described below are only some implementation manners of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0039] Figure 1 A flowchart of a method for adjusting the volume of a broadcast provided by the present application;

[0040] Figure 2 A structural schematic diagram of a broadcast volume adjustment method provided by the present invention;

[0041] Figure 3 The present invention provides a schematic structural diagram of an electronic device. DETAILED DESCRIPTION

[0042] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0043] Embodiment 1

[0044] Please refer to Figure 1 , is a flow chart of an embodiment of a broadcast volume adaptive adjustment method provided by an embodiment of the present invention, including steps S01-S04, each step is specifically as follows:

[0045] Step S01: obtaining a partition sound signal, and preprocessing the obtained partition sound signal to obtain partition sound data;

[0046] In this embodiment, high-quality audio sensors are set at reasonable positions in each partition, including but not limited to probes, microphone arrays, etc., to obtain sound signals in each partition. Accurate sound signal representation is provided by setting a suitable sampling frequency. In a preferred embodiment, a sampling frequency of 44.1kHZ or 48kHZ is selected, and the sampling bit number is set to 16 bits or 24 bits. The sound signals collected in real time by the audio sensor are preprocessed, providing a good basis for the subsequent separation of environmental noise and broadcast signals.

[0047] Preprocessing includes DC component removal, filtering, normalization, etc. DC component removal is achieved by subtracting the average value of the signal from the value of the sampling point, so that the average value of the signal is close to zero, avoiding the interference of DC component on subsequent frequency analysis and feature extraction. For example, in the station environment, there may be DC offset caused by equipment or environmental factors. Removing DC components can improve the accuracy of subsequent processing. Noise outside a specific frequency range is removed by filtering. For example, in the station environment, there is noise such as electromagnetic interference from equipment. These noises are removed by filtering to highlight the frequency range of broadcast signals and crowd noise.

[0048] Step S02: extracting features from the acquired sound data to obtain comprehensive features, separating the ambient noise and the broadcast signal based on the acquired comprehensive features, and measuring the corresponding average sound pressure levels of the separated ambient noise and the broadcast signal respectively.

[0049] In a preferred embodiment, by extracting comprehensive features, detailed features of the sound signal in frequency, time domain, and energy are obtained. These features provide detailed information for subsequent separation and classification, so that the classifier can accurately determine whether the sound signal belongs to a broadcast signal or noise; the extraction of comprehensive features can make subsequent classification and separation better adapt to different environments and achieve more effective separation.

[0050] First, the pre-processed sound data is subjected to fast Fourier transform, and the corresponding frequency domain coefficients are obtained by fast Fourier transform to represent the frequency and phase of different frequency components. The spectrum of the sound data, that is, the energy distribution of different frequency components, is obtained by fast Fourier transform analysis. The calculation formula of fast Fourier transform is: Among them, X[k] is the complex value at the corresponding frequency point in the frequency domain after fast Fourier transform, which contains the amplitude and phase information of the frequency component; x[n] represents the amplitude value of the original sound data at discrete time point n, k represents the frequency point index in the frequency domain, and j is an imaginary unit. is a complex exponential function, which plays a key role in converting time domain signals to frequency domain in fast Fourier transform. Determines the phase rotation angle of different frequency components.

[0051] The spectrum of the sound signal is obtained through fast Fourier transform, that is, the energy distribution of different frequency components. The main frequency is determined by finding the frequency component with the highest energy in the spectrum. In a preferred embodiment, the spectrum obtained through fast Fourier transform is |X[k]|, where |X[k]| represents the amplitude of the frequency domain coefficient. All frequency points are traversed to find the frequency value corresponding to the frequency point with the largest amplitude.

[0052] The calculation formula of main frequency is: Where f0 represents the main frequency, that is, the frequency component with the highest energy in the sound signal, and represents the index value corresponding to the frequency point with the largest amplitude in the spectrum after k0 is transformed by fast Fourier transform. s It represents the sampling frequency, that is, the number of times the sound signal is sampled per unit time. In a specific embodiment, the sampling frequency is set to 44100 Hz, which means that the sound signal is sampled 44100 times per second. N represents the length of the sound signal.

[0053] Determine the frequency range where the broadcast signal may appear, such as [f1, f2], and calculate the total energy in this frequency range. The calculation formula is: Among them, E b represents the sum of the energy of the broadcast signal in a specific frequency interval [f1, f2], k represents the frequency point index in the frequency, and the value range of k here is [f1, f2], that is, only the frequency points in the specific interval are considered. The total energy of the entire spectrum is calculated to determine the energy proportion of the broadcast signal in the frequency interval, where the calculation formula for the total spectrum energy is: Here, the value range of k is 0 to N-1, and the energy proportion of the broadcast signal is

[0054] The standard deviation of energy is calculated to measure the energy fluctuation. The calculation formula of the standard deviation is: When the energy standard deviation is small, it means that the energy fluctuation is small, which may be a broadcast signal. If the energy standard deviation is large, it means that the energy fluctuation is large, which may be noise.

[0055] For the collected signal x[n], the zero-crossing rate of the sound signal is calculated. The rhythm of the broadcast may cause the zero-crossing rate to change regularly within a certain time interval, while the zero-crossing rate of the ambient noise is usually more random. The zero-crossing rate calculation formula is: Where sgn(x) is a sign function, when x≥0, sgn(x)=1, when x<0, sgn(x)=-1; where ZCR represents the zero-crossing rate of the sound signal, N represents the length of the sound signal, and x[n] represents the amplitude value of the original sound signal at discrete time point n.

[0056] Specifically, the separation of environmental noise and broadcast signals based on the acquired comprehensive features specifically includes the following steps:

[0057] When the broadcasting system is not working, the subarea sound data is the first environmental noise of the corresponding subarea;

[0058] When the broadcasting system is working, the partition sound data is obtained in real time, and the extracted comprehensive features are input into a preset sound classification model to obtain classification information of the partition sound data;

[0059] For the part classified and marked as noise, the residual broadcast signal components are removed by filtering;

[0060] For the part classified and marked as broadcast signal, the residual noise component is removed by filtering;

[0061] The partitioned sound data and the corresponding classification labeling results are input into a separation algorithm, and the separated second environmental noise and the first broadcast signal are output.

[0062] Optionally, a random forest algorithm is used to construct a sound classification model, and the sound data with comprehensive features extracted in real time is organized into a format processed by the random forest algorithm. In a specific embodiment, there is a group of sound clips, each of which is represented by a feature vector, wherein the vector includes features such as the main frequency, the energy proportion of the frequency interval, and the energy standard deviation. The feature vectors are composed of a feature matrix X, and each sound clip has a corresponding label, which marks whether the clip belongs to environmental noise or a broadcast signal. These labels constitute the target vector y. Among them, the random forest is an integrated learning algorithm composed of multiple decision trees. First, the number of decision trees in the random forest is determined. For the training of each decision tree, a part of the samples are randomly selected from the feature matrix X and the target vector y with replacement to construct a sub-sample set. At the same time, when each node is split, a part of the features are randomly selected for consideration instead of using all the features, which can increase the diversity of the decision tree and reduce the risk of overfitting. Each decision tree starts from the root node and grows continuously according to the selected features and split conditions until certain stop conditions are met, such as reaching the maximum depth, the number of samples in the node is less than a certain threshold, etc.

[0063] For a newly received sound clip, its feature vector is input into the random forest after training. Each decision tree will classify the clip and provide the corresponding prediction result. The final classification result is determined by voting of all decision trees. When most decision trees believe that the clip is a broadcast signal, the clip is classified as a broadcast signal. Conversely, if most decision trees believe that it is environmental noise, it is classified as environmental noise.

[0064] According to the classification results, the sound clips are marked and stored in the corresponding data sets respectively. The mixed sound data are processed by the independent component analysis algorithm combined with the classification results of random forest as prior knowledge constraints to improve the accuracy of separation.

[0065] The mixed sound data is assisted in separation by using a filter. The broadcast signal is filtered using a bandpass filter for the data set marked as "broadcast", retaining only the components within the frequency range of the broadcast signal, removing the mixed noise components to obtain the first broadcast signal. For the data set marked as "noise", a bandstop filter can be used to block the broadcast signal components from passing through, retaining only the noise part to obtain the second environmental noise. The signal-to-noise ratio of the signal in the broadcast signal data set and the separated noise component is calculated. The higher the signal-to-noise ratio, the better the separation effect.

[0066] More specifically, the measuring the corresponding average sound pressure level specifically includes: dividing the second environmental noise and the first broadcast signal into time windows, and calculating the average sound pressure level corresponding to all time serial ports.

[0067] In a preferred embodiment, an initial time window size is first determined. This size can be selected based on a preliminary analysis of the sound signal. For example, a relatively small time window can be set so that the rapid changes of the signal can be captured more sensitively at the beginning. The separated second environmental noise and the first broadcast signal are analyzed in real time. For the second environmental noise, the total energy of the noise signal in each time window is calculated. If the energy changes greatly in several consecutive windows, it means that the noise has changed significantly during this period. The window size can be reduced for a more detailed analysis. If the energy changes in several adjacent windows exceed a preset change threshold, the window size is halved. The amplitude change of the noise signal is monitored. When the amplitude changes dramatically, it means that the noise situation may have changed significantly. The window size can also be reduced for better capture. For the first broadcast signal, the frequency stability of the broadcast signal is analyzed. When the frequency changes significantly in a short period of time, the window can be reduced to track the frequency dynamics. For example, when the main frequency or the main frequency range has a large offset in several consecutive windows, the window size can be adjusted. Taking into account the rhythm and periodicity of the broadcast signal, if the rhythm suddenly speeds up or slows down, the window size is adaptively adjusted to adapt to the signal change.

[0068] The root mean square value is calculated for the dynamically adjusted noise signal time window, where the root mean square (RMS) value is calculated as follows: Among them, n i is the sampling value of the noise signal in the corresponding window, k is the number of sampling points in the window, and the corresponding root mean square (RMS) value is converted into the sound pressure level, where the calculation formula is: As the window is dynamically adjusted, the calculation result of the sound pressure level of the noise signal is continuously updated. The sound pressure levels of all dynamically adjusted noise signal time windows are weighted averaged, and the weights can be determined according to the window size or the stability of the noise signal. For example, a higher weight is given to smaller windows because they can better reflect the rapid changes in noise. The calculation formula for the average sound pressure level is: Among them, L pni is the noise sound pressure level corresponding to multiple time windows, w ni is the corresponding weight.

[0069] For the dynamically adjusted broadcast signal time window, the root mean square (RMS) value is also calculated, where the root mean square (RMS) value is calculated as follows: Among them, b i is the sampling value of the noise signal in the corresponding window, l is the number of sampling points in the window, and the corresponding root mean square (RMS) value is converted into sound pressure level. The calculation formula is: With the dynamic adjustment of the window, the sound pressure level calculation result of the broadcast signal is continuously updated. The sound pressure levels of all windows of the broadcast signal are weighted averaged, where the average sound pressure level calculation formula of the broadcast signal is: Among them, L pbi is the corresponding broadcast signal sound pressure level in multiple time windows, w bi is the corresponding weight. Combined with dynamic window segmentation and signal separation, the average sound pressure levels of environmental noise such as noise and broadcast signals can be accurately measured; it should be noted that when the broadcast system is not working, the average sound pressure level of the first environmental noise can be directly obtained through the average sound pressure level formula.

[0070] Step S03: Based on the acquired environmental noise and average sound pressure level of the broadcast signal, the type of broadcast content, the collected environmental data and the reinforcement learning algorithm, adaptively adjust the optimal parameter strategy corresponding to the broadcast, including:

[0071] Defining the environmental state of the reinforcement learning algorithm based on the current environmental noise and the average sound pressure level of the broadcast signal, the type of broadcast content, and the collected environmental data; wherein the collected environmental data includes the flow of people in the partition and the time;

[0072] Defining an action space of the reinforcement learning algorithm according to broadcast parameters; wherein the action space includes: adjusting the volume, gain ratio and corresponding frequency response of the broadcast signal;

[0073] Determine a reward value corresponding to a specific execution action of a current broadcast signal parameter based on the environment state, the action space, and a reward function in the reinforcement learning algorithm;

[0074] Based on the reward value, the current parameter strategy of the broadcast signal is optimized to obtain the optimal parameter strategy.

[0075] In a specific embodiment, a Q value table is initialized for all possible combinations of environmental states and actions, wherein the environmental state includes multiple dimensions such as average sound pressure level of noise, broadcast content type, human flow level, time, etc., wherein the action space includes volume level, gain ratio value, frequency response type, etc.; initially all values ​​in the Q value table are set to a small random value; the environmental state and the factors and actions in the action space are quantized and encoded so as to be processed in the reinforcement learning algorithm. Among them, the influencing relationship indicators for determining the reward function may include broadcast intelligibility, signal-to-noise ratio, and adaptability to human flow and time.

[0076] More specifically, if the broadcast content is voice, the voice recognition software can be used to evaluate the proportion of words or sentences that the audience can correctly recognize under different broadcast parameters. Under specific actions, the higher the voice recognition accuracy, the higher the reward. When the broadcast content is important information, the broadcast signal must stand out from the ambient noise to ensure that the audience can hear it. This is reflected by calculating the signal-to-noise ratio of the broadcast signal and the ambient noise. The higher the signal-to-noise ratio, the more prominent the broadcast signal is relative to the ambient noise. Different signal-to-noise ratio thresholds can be set, and corresponding rewards can be given according to the actual signal-to-noise ratio. For example, when the signal-to-noise ratio is higher than the preset first threshold, a higher reward is given; when the signal-to-noise ratio is lower than the preset second threshold, a lower reward or penalty is given. At the same time, the corresponding signal-to-noise ratio threshold is set according to the type of broadcast content. For example, when the broadcast content is music, the corresponding thresholds are the third threshold and the fourth threshold respectively; when there is a large flow of people, the ambient noise usually increases, and the broadcast volume needs to be increased or other parameters need to be adjusted to ensure the audibility of the broadcast; different rewards are given according to the level of traffic. For example, when the flow of people changes from low to high, the reward value is gradually increased to encourage the algorithm to increase the broadcast volume or gain ratio; the environmental noise and people's sensitivity to broadcasting may be different in different time periods. For example, during the day, the environmental noise may be high and a higher volume is required; while at night, people may want the broadcast volume to be lower to avoid disturbing others. Corresponding rewards can be given according to the time category. For example, in a specific time period, such as at night, lowering the volume can be rewarded, while raising the volume during the day can be rewarded. The reward function R can be expressed by the following formula:

[0077] R=w1×C+w2×S+w3×P+w4×T;

[0078] Where R is the total reward value, C is the reward related to the signal-to-noise ratio, S is the reward related to broadcast intelligibility, P is the reward related to traffic, and T is the reward related to time; w1, w2, w3, and w4 are the corresponding weight coefficients, which are used to adjust the importance of each factor in the reward function; the weight coefficient can be dynamically adjusted according to the actual situation. For example, in a certain period of time, broadcast intelligibility becomes more important, and the weight of the relevant reward can be increased. According to changes in the environment and needs, the parameters of the reward function are updated to adapt to the dynamic environment. When determining the reward function, consider the balance between multiple objectives, and use multi-objective optimization techniques to find the best compromise between different objectives, such as Pareto technology.

[0079] On this basis, a specific execution action is randomly selected from the spatial actions as the starting action, and a greedy algorithm is used for balanced exploration. The selected starting action is executed and the new environmental state generated after the execution of the starting action is observed in real time. The reward value after each specific execution action in the action space is calculated according to the set reward function. Considering the impact of each specific execution action in the action space on the environmental state, risk penalties or rewards can be added when updating the reward value Q, and the Q value is updated by the following formula:

[0080] Q(s,a)=Q(s,a)+α*(r+γ*max(Q(s',a'))-Q(s,a));

[0081] Among them, Q(s,a) represents the reward value of taking action a in state s, which represents the estimate of the long-term expected cumulative reward for the specific execution action taken under the environmental state; α represents the learning rate; r represents the immediate reward obtained after executing action a; γ represents the discount factor, which is used to measure the importance of future rewards; max(Q(s',a')) represents the Q value of the action with the largest Q value among all possible actions a' in the new state s', which represents the expected reward estimate for the best future action; on this basis, by continuously updating the reward value Q until the reward value Q stabilizes, the optimal parameter strategy for the broadcast signal is output. When the environmental state changes, the volume, gain ratio and frequency response of the broadcast signal are adjusted according to the optimal strategy determined by the Q value table to achieve adaptive broadcast parameter adjustment.

[0082] S04: adjusting the current parameters of the broadcast signal to the parameters corresponding to the optimal parameter strategy based on the optimal parameter strategy, including:

[0083] Sending the optimal parameter strategy to the broadcast to be adjusted, and determining the volume, gain ratio and frequency response corresponding to the broadcast to be adjusted in the optimal parameter strategy;

[0084] Based on the dynamic adjustment algorithm, the current parameters of the broadcast to be adjusted are adjusted to the corresponding parameters in the optimal parameter strategy.

[0085] It should be noted that after the adjusted broadcast receives the optimal parameter strategy, it needs to parse the strategy information to determine the volume, gain ratio and frequency response to be used for itself. For example: the strategy indicates that the broadcast in a specific area should set the volume to a medium level (for example, assuming the volume range is 0-10, the medium level is 5) and the gain ratio to 1.5 in the current environment, and enhance the frequency response of the mid- and low-frequency bands to improve voice clarity.

[0086] The dynamic adjustment algorithm aims to smoothly adjust the current parameters of the broadcast to the parameters specified by the optimal parameter strategy to avoid sudden changes that may cause discomfort to the audience. In this embodiment, a method based on linear interpolation is adopted, and each adjustment increases or decreases an increment △P, which is gradually increased within a certain period of time; the parameters of the broadcast signal are effectively adjusted according to the optimal parameter strategy to adapt to different environments and broadcasting needs, thereby improving the broadcast effect and audience experience.

[0087] The embodiments of the present invention provide a method, device, equipment, medium and product for adaptively adjusting broadcast volume. The method obtains sound signals in different partitions, pre-processes the acquired partition sound signals to obtain partition sound data, extracts features from the acquired sound data to obtain comprehensive features, and can more effectively separate noise and broadcast based on the comprehensive features. The corresponding average sound pressure levels of the separated environmental noise and broadcast signals are respectively measured, and the sound pressure level data is used as the basis for adjusting broadcast parameters and environmental control to ensure the quality of sound propagation and help evaluate the propagation effect of broadcast signals. The corresponding optimal parameter strategy is determined by a reinforcement learning algorithm based on the acquired environmental noise in the partition and the average sound pressure level of the broadcast signal, the type of broadcast content and other data, and the broadcast parameters are adaptively and dynamically adjusted based on the dynamic adjustment technology to ensure that the audience's auditory experience is met in a complex acoustic environment, ensure the best propagation effect of the broadcast signal under different circumstances, and ensure the audience's acquisition of key information.

[0088] Embodiment 2

[0089] Please refer to Figure 2 , is a schematic diagram of the structure of a broadcast volume adaptive adjustment device provided in an embodiment of the present application. The device can be implemented by software and / or hardware, and is generally integrated on any electronic device with network communication function, including but not limited to: servers, computers, personal digital assistants and other devices. Figure 2 The embodiment of the present application provides a broadcast volume adaptive adjustment device, including: a real-time acquisition module 201, a separation measurement module 202, a strategy determination module 203 and a broadcast adjustment module 204;

[0090] The real-time acquisition module 201 is used to obtain partition sound signals and pre-process the obtained partition sound signals to obtain partition sound data;

[0091] The separation and measurement module 202 extracts features from the acquired sound data to obtain comprehensive features, separates the ambient noise and the broadcast signal based on the acquired comprehensive features, and measures the corresponding average sound pressure levels of the separated ambient noise and the broadcast signal respectively;

[0092] The strategy determination module 203 is used to adaptively adjust the optimal parameter strategy corresponding to the broadcast according to the acquired environmental noise and average sound pressure level of the broadcast signal, the broadcast content type, the collected environmental data and the reinforcement learning algorithm;

[0093] The broadcast adjustment module 204 is used to adjust the current parameters of the broadcast signal to the parameters corresponding to the optimal parameter strategy according to the optimal parameter strategy.

[0094] Embodiment 3

[0095] Figure 3 A schematic diagram of the structure of an electronic device provided for an embodiment of the present invention. The electronic device 10 is intended to represent various forms of digital computers, such as laptop computers, desktop computers, workstations, personal digital assistants, servers, blade servers, mainframe computers, and other suitable computers. The electronic device may also represent various forms of mobile devices, such as personal digital processing, cellular phones, smart phones, wearable devices (such as helmets, glasses, watches, etc.) and other similar computing devices. The components shown herein, their connections and relationships, and their functions are merely examples and are not intended to limit the implementation of the present invention described and / or required herein.

[0096] like Figure 3 As shown, the electronic device 10 includes at least one processor 11, and a memory connected to the at least one processor 11, such as a read-only memory (ROM) 12, a random access memory (RAM) 13, etc., wherein the memory stores a computer program that can be executed by at least one processor, and the processor 11 can perform various appropriate actions and processes according to the computer program stored in the read-only memory (ROM) 12 or the computer program loaded from the storage unit 18 to the random access memory (RAM) 13. In the RAM 13, various programs and data required for the operation of the electronic device 10 can also be stored. The processor 11, the ROM 12, and the RAM 13 are connected to each other through a bus 14. An input / output (I / O) interface 15 is also connected to the bus 14.

[0097] A number of components in the electronic device 10 are connected to the I / O interface 15, including: an input unit 16, such as a keyboard, a mouse, etc.; an output unit 17, such as various types of displays, speakers, etc.; a storage unit 18, such as a disk, an optical disk, etc.; and a communication unit 19, such as a network card, a modem, a wireless communication transceiver, etc. The communication unit 19 allows the electronic device 10 to exchange information / data with other devices through a computer network such as the Internet and / or various telecommunication networks.

[0098] The processor 11 may be a variety of general and / or special processing components with processing and computing capabilities. Some examples of the processor 11 include, but are not limited to, a central processing unit (CPU), a graphics processing unit (GPU), various special artificial intelligence (AI) computing chips, various processors running machine learning model algorithms, a digital signal processor (DSP), and any appropriate processor, controller, microcontroller, etc. The processor 11 executes the various methods and processes described above, such as the automatic frequency tuning method of a wireless microphone.

[0099] In some embodiments, the broadcast volume automatic adjustment method can be implemented as a computer program, which is tangibly contained in a computer-readable storage medium, such as a storage unit 18. In some embodiments, part or all of the computer program can be loaded and / or installed on the electronic device 10 via the ROM 12 and / or the communication unit 19. When the computer program is loaded into the RAM 13 and executed by the processor 11, one or more steps of the broadcast volume automatic adjustment method described above can be performed. Alternatively, in other embodiments, the processor 11 can be configured to execute the automatic frequency tuning method of the wireless microphone in any other appropriate manner (for example, by means of firmware).

[0100] Various implementations of the systems and techniques described above herein can be implemented in digital electronic circuit systems, integrated circuit systems, field programmable gate arrays (FPGAs), application specific integrated circuits (ASICs), application specific standard products (ASSPs), systems on chips (SOCs), load programmable logic devices (CPLDs), computer hardware, firmware, software, and / or combinations thereof. These various implementations can include: being implemented in one or more computer programs that can be executed and / or interpreted on a programmable system including at least one programmable processor, which can be a special purpose or general purpose programmable processor that can receive data and instructions from a storage system, at least one input device, and at least one output device, and transmit data and instructions to the storage system, the at least one input device, and the at least one output device.

[0101] Computer programs for implementing the methods of the present invention may be written in any combination of one or more programming languages. These computer programs may be provided to a processor of a general-purpose computer, a special-purpose computer, or other programmable data processing device, so that when the computer program is executed by the processor, the functions / operations specified in the flow chart and / or block diagram are implemented. The computer program may be executed entirely on the machine, partially on the machine, partially on the machine and partially on a remote machine as a stand-alone software package, or entirely on a remote machine or server.

[0102] In the context of the present invention, a computer-readable storage medium may be a tangible medium that may contain or store a computer program for use by or in combination with an instruction execution system, device or equipment. A computer-readable storage medium may include, but is not limited to, electronic, magnetic, optical, electromagnetic, infrared, or semiconductor systems, devices or equipment, or any suitable combination of the foregoing. Alternatively, a computer-readable storage medium may be a machine-readable signal medium. A more specific example of a machine-readable storage medium may include an electrical connection based on one or more lines, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.

[0103] To provide interaction with a user, the systems and techniques described herein may be implemented on an electronic device having: a display device (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor) for displaying information to the user; and a keyboard and a pointing device (e.g., a mouse or trackball) through which the user can provide input to the electronic device. Other types of devices may also be used to provide interaction with the user; for example, the feedback provided to the user may be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user may be received in any form (including acoustic input, voice input, or tactile input).

[0104] The systems and techniques described herein may be implemented in a computing system that includes backend components (e.g., as a data server), or a computing system that includes middleware components (e.g., an application server), or a computing system that includes frontend components (e.g., a user computer with a graphical user interface or a web browser through which a user can interact with implementations of the systems and techniques described herein), or a computing system that includes any combination of such backend components, middleware components, or frontend components. The components of the system may be interconnected by any form or medium of digital data communication (e.g., a communication network). Examples of communication networks include: a local area network (LAN), a wide area network (WAN), a blockchain network, and the Internet.

[0105] A computing system may include a client and a server. The client and the server are generally remote from each other and usually interact through a communication network. The client and server relationship is generated by computer programs running on the corresponding computers and having a client-server relationship with each other. The server may be a cloud server, also known as a cloud computing server or cloud host, which is a host product in the cloud computing service system to solve the defects of difficult management and weak business scalability in traditional physical hosts and VPS services.

[0106] Embodiment 4

[0107] The present application embodiment also provides a kind of computer program product, including a computer program, which implements the broadcast volume adaptive adjustment method provided by any embodiment of the present application when executed by a processor. In the process of implementation, the computer program product can be written in one or more programming languages ​​or a combination thereof to perform the computer program code for the operation of the present invention, and the programming language includes an object-oriented programming language, such as Java, Smalltalk, C++, and also includes a conventional procedural programming language, such as "C" language or similar programming language. The program code can be executed completely on the user's computer, partially on the user's computer, as an independent software package, partially on the user's computer, partially on the remote computer, or completely on the remote computer or server. In the case of a remote computer, the remote computer can be connected to the user's computer through any kind of network, including a local area network (LAN) or a wide area network (WAN), or it can be connected to an external computer (for example, using an Internet service provider to connect through the Internet).

[0108] It should be understood that the various forms of processes shown above can be used to reorder, add or delete steps. For example, the steps described in the present invention can be executed in parallel, sequentially or in different orders, as long as the desired results of the technical solution of the present invention can be achieved, and this document does not limit this.

[0109] The specific embodiments described above further illustrate the purpose, technical solutions and beneficial effects of the present invention. It should be understood that the above description is only a specific embodiment of the present invention and is not intended to limit the scope of protection of the present invention. It is particularly pointed out that for those skilled in the art, any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention should be included in the scope of protection of the present invention.

Claims

1. A method for adaptively adjusting broadcast volume, characterized in that: include: Acquire partition sound signals, and pre-process the acquired partition sound signals to acquire partition sound data; Extracting features from the acquired sound data to obtain comprehensive features, separating the ambient noise and the broadcast signal based on the acquired comprehensive features, and measuring the corresponding average sound pressure levels of the separated ambient noise and the broadcast signal respectively; Adaptively adjust the optimal parameter strategy corresponding to the broadcast based on the acquired environmental noise and average sound pressure level of the broadcast signal, the type of broadcast content, the collected environmental data and the reinforcement learning algorithm; Based on the optimal parameter strategy, current parameters of the broadcast signal are adjusted to parameters corresponding to the optimal parameter strategy.

2. A method for adaptively adjusting broadcast volume according to claim 1, characterized in that: The comprehensive features include spectrum features, time domain features and sound signal energy features.

3. A broadcast volume adjustment method according to claim 1, characterized in that: The separation of the environmental noise and the broadcast signal based on the obtained comprehensive features specifically includes the following steps: When the broadcasting system is not working, the subarea sound data is the first environmental noise of the corresponding subarea; When the broadcasting system is working, the partition sound data is obtained in real time, and the extracted comprehensive features are input into a preset sound classification model to obtain classification information of the partition sound data; For the part classified and marked as noise, the residual broadcast signal components are removed by filtering; For the part classified and marked as broadcast signal, the residual noise component is removed by filtering; The partitioned sound data and the corresponding classification labeling results are input into a separation algorithm, and the separated second environmental noise and the first broadcast signal are output.

4. A broadcast volume adjustment method according to claim 1, characterized in that: The measuring corresponding average sound pressure level specifically includes: dividing the second environmental noise and the first broadcast signal into time windows, and calculating the average sound pressure level corresponding to all time serial ports.

5. A broadcast volume adjustment method according to claim 1, characterized in that: The adaptive adjustment of the optimal parameter strategy corresponding to the broadcast based on the acquired environmental noise and average sound pressure level of the broadcast signal, the broadcast content type, the collected environmental data and the reinforcement learning algorithm includes: Defining the environmental state of the reinforcement learning algorithm based on the current environmental noise and the average sound pressure level of the broadcast signal, the type of broadcast content, and the collected environmental data; wherein the collected environmental data includes the flow of people in the partition and the time; Defining an action space of the reinforcement learning algorithm according to broadcast parameters; wherein the action space includes: adjusting the volume, gain ratio and corresponding frequency response of the broadcast signal; Determine a reward value corresponding to a specific execution action of a current broadcast signal parameter based on the environment state, the action space, and a reward function in the reinforcement learning algorithm; Based on the reward value, the current parameter strategy of the broadcast signal is optimized to obtain the optimal parameter strategy.

6. A broadcast volume adjustment method according to claim 1, characterized in that: The adjusting the current parameters of the broadcast signal to the parameters corresponding to the optimal parameter strategy based on the optimal parameter strategy includes: Sending the optimal parameter strategy to the broadcast to be adjusted, and determining the volume, gain ratio and frequency response corresponding to the broadcast to be adjusted in the optimal parameter strategy; Based on the dynamic adjustment algorithm, the current parameters of the broadcast to be adjusted are adjusted to the corresponding parameters in the optimal parameter strategy.

7. A broadcast volume adjustment device, characterized in that: It includes a real-time acquisition module, a separation measurement module, a strategy determination module and a broadcast adjustment module; The real-time acquisition module is used to obtain partition sound signals and pre-process the obtained partition sound signals to obtain partition sound data; The separation and measurement module extracts features from the acquired sound data to obtain comprehensive features, separates the ambient noise and the broadcast signal based on the acquired comprehensive features, and measures the corresponding average sound pressure levels of the separated ambient noise and the broadcast signal respectively; The strategy determination module is used to adaptively adjust the optimal parameter strategy corresponding to the broadcast according to the acquired environmental noise and average sound pressure level of the broadcast signal, the type of broadcast content, the collected environmental data and the reinforcement learning algorithm; The broadcast adjustment module is used to adjust the current parameters of the broadcast signal to the parameters corresponding to the optimal parameter strategy according to the optimal parameter strategy.

8. An electronic device, characterized in that: include: one or more processors; A storage device for storing one or more programs; When the one or more programs are executed by the one or more processors, the one or more processors implement a broadcast volume adjustment method as described in any one of claims 1-6.

9. A storage medium containing computer executable instructions, characterized in that: The computer executable instructions are used to execute a broadcast volume adjustment method as described in any one of claims 1-6 when executed by a computer processor.

10. A computer program product, comprising a computer program, wherein when the computer program is executed by a processor, the computer program implements the broadcast volume adjustment method according to any one of claims 1 to 6.

Images