A method, device, equipment and medium for generating electric vehicle throttle sound linkage

Abstract

The present invention relates to the technical field of electric vehicles, and discloses a method, device, equipment and medium for generating linkage of electric vehicle throttle sound effects, which obtains electric vehicle status information, obtains matching engine audio according to the electric vehicle status information, synthesizes simulated throttle sound effects according to the matching engine audio, and controls a power amplifier to drive a speaker to output simulated throttle sound effects according to the simulated throttle sound effects. Sound effect feedback during driving or riding is achieved, which has a high matching degree with the electric vehicle status information and high fidelity, and enhances the driver's experience and driving feeling.

Description

Technical Field

[0001] The present invention relates to the technical field of electric vehicles, and in particular to a method, device, equipment and medium for generating a linkage sound effect of a throttle of an electric vehicle. Background Art

[0002] With the development trend of energy conservation and emission reduction around the world, the new energy vehicle industry has developed rapidly. Pure electric vehicles have flooded into the market, and the drive motor has replaced the traditional engine. The overall noise of the vehicle has decreased, but the NVH problem (Noise, Vibration, Harshness, which is a comprehensive issue to measure the quality of automobile manufacturing) faces new challenges. The interior of a pure electric vehicle is relatively quiet when driving, and it cannot provide the driver with sound feedback reflecting the changes in the vehicle status. Summary of the invention

[0003] In view of this, the present invention provides a method for generating a linkage sound effect of an electric vehicle throttle, which solves the technical problem that the sound of the prior art electric vehicles is relatively quiet when running, and cannot provide the driver or rider with sound feedback reflecting changes in the vehicle state, resulting in a poor driving experience.

[0004] According to a first aspect of the present invention, a method for generating a linkage sound effect of a throttle of an electric vehicle is provided, comprising:

[0005] Acquiring electric vehicle status information, wherein the electric vehicle status information includes electric vehicle acceleration control state, electric vehicle motor speed and electric vehicle acceleration;

[0006] According to the electric vehicle state information, obtaining matching engine audio;

[0007] Synthesize simulated throttle sound effects based on matching engine audio;

[0008] According to the simulated throttle sound effect, the power amplifier is controlled to drive the speaker to output the simulated throttle sound effect.

[0009] According to a second aspect of the present invention, a method and device for generating a linkage sound effect of an electric vehicle throttle is provided, comprising:

[0010] An acquisition module, used to acquire electric vehicle status information, wherein the electric vehicle status information includes electric vehicle acceleration control state, electric vehicle motor speed and electric vehicle acceleration;

[0011] A query module, used for acquiring matching engine audio according to the electric vehicle status information;

[0012] Synthesis module, used to synthesize simulated throttle sound effects based on matching engine audio;

[0013] The output module is used to control the power amplifier to drive the speaker to output the simulated throttle sound effect according to the simulated throttle sound effect.

[0014] According to a third aspect of the present invention, there is provided a computer device comprising a memory, a processor and a computer program stored in the memory and executable on the processor, wherein the processor implements the steps of the above-mentioned method for generating linkage sound effects of electric vehicle throttles when executing the computer program.

[0015] According to a fourth aspect of the present invention, there is provided a computer-readable storage medium storing a computer program, which, when executed by a processor, implements the steps of the above-mentioned method for generating linkage sound effects of electric vehicle throttles.

[0016] Through the above technical solution, the present invention provides a method, device, equipment and medium for generating linkage of electric vehicle throttle sound effects, which obtains electric vehicle status information, obtains matching engine audio according to the electric vehicle status information, synthesizes simulated throttle sound effects according to the matching engine audio, and controls the power amplifier to drive the speaker to output simulated throttle sound effects according to the simulated throttle sound effects. Sound effect feedback during driving or riding is achieved, which has a high matching degree and high fidelity with the electric vehicle status information, and enhances the driver's experience and driving feeling.

[0017] The above description is only an overview of the technical solution of the present invention. In order to more clearly understand the technical means of the present invention and implement it according to the contents of the specification, and to make the above and other purposes, features and advantages of the present invention more obvious and easy to understand, the specific implementation methods of the present invention are listed below. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] The drawings described herein are used to provide a further understanding of the present invention and constitute a part of the present invention. The exemplary embodiments of the present invention and their descriptions are used to explain the present invention and do not constitute an improper limitation of the present invention. In the drawings:

[0019] Figure 1 A schematic diagram showing an application scenario of a method for generating a linkage sound effect of an electric vehicle throttle provided in an embodiment of the present invention;

[0020] Figure 2 A flow chart of a method for generating electric vehicle throttle sound linkage provided in the first embodiment of the present invention is shown;

[0021] Figure 3 A schematic diagram of the synthesis of the switch sound effect provided in the first embodiment of the present invention is shown;

[0022] Figure 4 A flow chart of a method for generating electric vehicle throttle sound linkage provided in a second embodiment of the present invention is shown;

[0023] Figure 5 A structural schematic diagram of a method and device for generating linkage sound effects of electric vehicle throttles provided in an embodiment of the present invention is shown. DETAILED DESCRIPTION

[0024] The specific implementation of the present invention will be described in detail below with reference to the accompanying drawings and in combination with embodiments. It should be noted that the embodiments and features in the embodiments of the present invention can be combined with each other without conflict.

[0025] The electric vehicle throttle sound effect linkage generation method provided by the embodiment of the present invention can be applied to Figure 1 The hardware system can be installed in a variety of transportation tools, such as bicycles, electric bicycles, motorcycles, motor vehicles, etc. The hardware system can share hardware resources with the original system, collect throttle opening signals, motor speed signals, and motor torque signals through a bus module (such as CAN bus, electric bicycle bus general technical specification standard is QB / T 5869-2023), and obtain electric vehicle status information through a digital processor (such as a digital processing chip) according to a method for generating electric vehicle electric door sound effects linkage provided by an embodiment of the present invention for the collected data variables, obtain matching engine audio according to the electric vehicle status information, and synthesize simulated electric door sound effects according to the matching engine audio. Then, according to the simulated electric door sound effects, the sound effect control module controls the power amplifier to drive the speaker to output the simulated electric door sound effects. For four-wheeled vehicles, the simulated electric door sound effects need to distinguish between built-in sound effects and external sound effects, and play the simulated electric door sound effects respectively. For two-wheeled vehicles, only external sound effects can be played.

[0026] The present invention is described in detail below through specific embodiments.

[0027] Embodiment 1

[0028] like Figure 2 As shown, a method for generating electric vehicle throttle sound linkage provided in an embodiment of the present invention includes:

[0029] Step 201, obtaining electric vehicle status information;

[0030] Among them, the electric vehicle status information includes the electric vehicle acceleration control status, the electric vehicle motor speed and the electric vehicle acceleration. For four-wheeled vehicles, the above data can be collected through the CAN bus, and for electric bicycles, the electric bicycle bus general technical specification standard QB / T 5869-2023 can be used to collect the above data.

[0031] Step 202: Acquire matching engine audio according to the electric vehicle status information;

[0032] Among them, the segmented audio sample library is queried according to the parameters of the electric vehicle status information to obtain segmented engine audio, such as the motor speed range of 1000-1500RPM, and the motor speed is used as an index to match 10 segmented audios arranged in chronological order, and the subsequent steps are synthesized based on these segmented audios. In addition, it should be noted that users can also match based on other parameters, such as vehicle type, sound effect mode (dynamic, science fiction, natural, turbo, etc.).

[0033] Step 203, synthesizing a simulated throttle sound effect according to the matched engine audio;

[0034] Step 203 specifically includes:

[0035] Step 203-1, splicing the engine audio sequence obtained from the segmented audio sample library according to the preset overlapping time interval to generate spliced ​​audio;

[0036] Among them, since the continuous playback of sound relies on the reading of sound fragments and the superposition between the fragments, and since the main orders of the sound fragments in the sound sample database all have the same initial phase, the frequency and amplitude of the sound data in the overlapping superposition area within the time domain are changed by using the time domain synchronous superposition method, so as to achieve the increase and decrease of the frequency, the smooth and continuous amplitude, and the synchronous connection of the frequency and amplitude curves of the adjacent superimposed short-time signals. The sound fragments corresponding to the rotation speed are retrieved from the extracted sound fragment database, and they are overlapped and added in turn. The overlapping and addition of the sound samples makes the amplitude of the adjacent sound fragments smooth and continuous. The overlapping length set in the embodiment of the present invention needs to be long enough to cover the amplitude and frequency differences between the sound samples during the synthesis process.

[0037] Step 203-2, determining the audio gain coefficient according to the opening degree of the acceleration control state of the electric vehicle;

[0038] Among them, this step uses the accelerator pedal opening signal to adjust the synthesized sound amplitude by a modulation coefficient of α, and finally obtains a more realistic synthesized engine sound. It should be noted that the opening signal here depends on the control amount of the accelerator. For four-wheeled vehicles, it is controlled by the accelerator pedal opening signal, while for two-wheeled vehicles, the synthesized sound amplitude is adjusted by a modulation coefficient of α through the rotation opening signal of the accelerator handle. Figure 3 The figure shows the throttle pedal opening signal corresponding to different modulation coefficients, and the schematic diagram of multi-segment audio synthesis after amplitude transformation of the segmented audio. In the figure, throttle opening 1 corresponds to segmented audio 1 and modulation coefficient α1, throttle opening 2 corresponds to segmented audio 2 and modulation coefficient α2, and throttle opening 3 corresponds to segmented audio 3 and modulation coefficient α3.

[0039] Step 203-3: adjust the amplitude of the spliced ​​audio according to the audio gain coefficient to synthesize a simulated switch sound effect.

[0040] Among them, Figure 3 As shown, multiple audio segments after amplitude modulation are superimposed according to the preset overlapping parts to synthesize the simulated switch sound effect.

[0041] Step 204: According to the simulated throttle sound effect, control the power amplifier to drive the speaker to output the simulated throttle sound effect.

[0042] Before step 202, the following steps are included:

[0043] Step 202-a, collecting audio of the sound-generating component to be simulated in different working states;

[0044] Among them, the sound effects to be simulated may include various forms such as fuel engines, aircraft turbine engines, rocket launch sound effects, etc. In the actual sound effect collection process, the corresponding audio files can be flexibly collected according to the needs and the corresponding different working states can be manually defined. For example, the engine and exhaust pipe sounds can be simulated under acceleration to simulate the sound wave sound effect, simulate the wind-breaking acceleration sound effect, etc. For the engine, the working state includes the idling state, the acceleration state and the deceleration state.

[0045] Step 202-b, segmenting the collected audio according to different preset speed ranges;

[0046] The sound samples of the corresponding input signal are extracted by using a time window of appropriate length. For the engine noise model, each sound sample must contain an engine cycle and some overlapping information to ensure a smooth transition between sound samples during the synthesis process. The collected full-speed engine audio is divided into small sound clips. For some special sound effects such as artificially synthesized sound effects, aircraft turbine engines, rocket launch sound effects, etc., since they cannot be recorded in practice, they can be defined by artificial segmentation according to the audio to correspond to different preset speed ranges.

[0047] Step 202-c, performing short-time Fourier transform on the segmented audio to obtain a frequency spectrum distribution corresponding to the segmented audio;

[0048] Among them, short-time Fourier transform realizes audio order analysis, extracts the main order of audio, decomposes and processes the entire collected original sound signal, and forms a series of sound signal fragments. These fragments are arranged from low to high according to the main order frequency, and the main order components in each fragment have the same initial phase and superposition area.

[0049] Step 202-d: filtering the segmented audio according to the frequency spectrum distribution, extracting frequency components with energy proportion greater than 80%, and generating audio samples;

[0050] Among them, the engine sound is mainly low-frequency, which can give people a strong sense of power, while the high-frequency sound gives people an uncomfortable feeling of irritability, giving people a better sound quality. After repeated audition and subjective evaluation and comprehensive consideration of various aspects, the original sound signal is filtered using an FIR low-pass filter. The low-pass filter uses window functions including Rectangle window, Hanning window, Hamming window, Blackman window, Kaiser window, etc. The optimal cutoff frequency is set to 1000Hz.

[0051] Step 201 - e: store the audio samples corresponding to the rotation speed of the fuel engine to be simulated, and generate a segmented audio sample library.

[0052] The audio samples of each segment are numbered sequentially, corresponding to a certain range of accelerator pedal opening values, and the engine acceleration sound is synthesized using the idea of ​​waveform splicing.

[0053] Because the speed of the engine sound sample used cannot completely cover the motor speed (usually the motor speed is greater than the fuel engine speed), it is necessary to convert the speed corresponding to the engine sound sample into the motor speed. Therefore, step 201-b5 includes:

[0054] Step 201-e1, obtaining the speed range of the electric vehicle motor and the speed range of the fuel engine to be simulated;

[0055] Step 201-e2, converting the speed of the fuel engine to be simulated corresponding to the audio sample into the speed of the electric vehicle motor according to the proportionality coefficient between the speed range of the electric vehicle motor and the speed range of the fuel engine to be simulated;

[0056] Among them, usually the motor speed range is larger than the speed range of the engine to be simulated, so proportional mapping is required. For example, the speed range of the engine to be simulated is 0-5000RPM, and the motor speed of the electric vehicle is 0-8000RPM. The audio segment samples are indexed and stored according to the speed range of the engine to be simulated. Therefore, it is necessary to convert the motor speed of the electric vehicle into the speed of the engine to be simulated based on the speed ratio coefficient (5000 / 8000).

[0057] Step 201 - e3, storing the audio samples and the converted electric vehicle motor speeds in correspondence to generate a segmented audio sample library.

[0058] In practical applications, digital modeling can be used to record synthesized sound effects. For example, digital synthesis technology can be used to record the sound of the motor itself rotating and the sound of the motor breaking through the wind during the high-speed operation of a high-speed electric motorcycle. Then, an audio sample library can be synthesized in accordance with the sound synthesis modeling method, and the collected sound can be preprocessed, including denoising and enhancement, to improve the sound quality. The processed sound is analyzed to extract characteristic parameters, such as frequency, amplitude, duration, etc., and then a sound model is established based on the extracted characteristic parameters to simulate the sound of an electric motorcycle. The sound model is used to synthesize the required sound effects, including the sound of an electric motorcycle at different speeds. The synthesized sound is evaluated to ensure that it is highly similar to the actual electric motorcycle sound, such as synthesizing sound effects in the styles of anime, two-dimensional, technology, punk, etc. According to the evaluation results, the sound synthesis technology is adjusted and optimized to improve the sound quality and authenticity.

[0059] The embodiment of the present invention provides a method for generating electric vehicle throttle sound effects linkage, which obtains electric vehicle status information, obtains matching engine audio according to the electric vehicle status information, synthesizes simulated throttle sound effects according to the matching engine audio, and controls the power amplifier to drive the speaker to output the simulated throttle sound effects according to the simulated throttle sound effects. Sound effect feedback during driving or riding is achieved, which has a high degree of matching with the electric vehicle status information and high fidelity, and enhances the driver's experience and driving feeling.

[0060] Embodiment 2

[0061] In order to better simulate the sound effect of the electric vehicle throttle and improve the fidelity of the simulation with the actual fuel engine throttle working state, the electric vehicle state information is further divided into idle state, deceleration state and acceleration state in embodiment 2 of the present invention, and then the appropriate sound effect is matched according to the working state and motor speed.

[0062] Based on the first embodiment of the present invention, Figure 4 As shown, step 202 may specifically include

[0063] Step 202-1: when it is determined that the current acceleration is greater than zero and the acceleration control state of the electric vehicle is in a failed state according to the electric vehicle state information, it is determined that the engine is in an idle state, and according to the electric vehicle motor speed, an engine audio sequence matching the idle state is obtained from the segmented audio sample library;

[0064] Step 202-2: when it is determined according to the electric vehicle state information that the current acceleration is less than zero and the electric vehicle acceleration control state is a braking state, it is determined that the engine is in a deceleration state, and based on the electric vehicle motor speed, an engine audio sequence matching the deceleration state is obtained from the segmented audio sample library;

[0065] Step 202-3: when it is determined according to the electric vehicle state information that the current acceleration is greater than zero and the electric vehicle acceleration control state is the acceleration state, the engine is determined to be in the acceleration state, and according to the electric vehicle motor speed, an engine audio sequence matching the acceleration state is obtained from the segmented audio sample library.

[0066] It should be noted that in the process of creating the segmented audio sample library in step 201, it is necessary to distinguish different state information for audio collection and add the state information for corresponding storage. The other processes are the same as those introduced in the first embodiment of the present invention and will not be repeated here.

[0067] Embodiment 3

[0068] For a four-wheel vehicle, the sound effect environment inside and outside the vehicle is different, and it is necessary to distinguish different sound effects inside and outside the vehicle. For example, it is possible that the throttle sound effect needs to be simulated inside the vehicle while the vehicle needs to remain quiet. In this case, it is necessary to perform sound effect simulations inside and outside the vehicle respectively. Step 201-a in the first embodiment of the present invention may specifically include:

[0069] The audio signals of the fuel engine to be simulated in different working states are collected from inside and outside the vehicle respectively;

[0070] The corresponding step 202 specifically includes obtaining the matched in-vehicle engine audio and out-vehicle engine audio respectively according to the electric vehicle state information.

[0071] Embodiment 4

[0072] Usually, there is a response delay in obtaining the electric vehicle status information in step 201, and the system program delay accounts for a large proportion of the response delay. For example: the embedded system delay is determined by the circuit, hardware structure design and algorithm of the device itself, and the time consumed in processing and outputting signals is the system delay. Among them, the algorithm processing signal time has a greater impact, such as filtering and reading the CAN signal emitted by the vehicle in real time, audio retrieval time, audio sampling time, decoding time and audio processing. The interval delay is the interval time between the system program entering the CAN signal recognition and the algorithm identifying the vehicle working condition to index the audio data after the previous simulated sound audio sound is finished, and then starting to play the next simulated sound effect. The embodiment of the present invention divides the simulated sound effect into multiple segments, and then retrieves the corresponding sound segment through the index to simulate the sound. After the previous sound effect of the program system is finished, the DAC module is in a short silent stage without sound playback until the next sound is indexed and called out. If the silent interval exceeds the 50ms perception threshold of human perception of sound intervals, the user will acoustically perceive the discontinuity of the sound. To solve this technical problem, in the fourth embodiment of the present invention, the corresponding sound effect clips can be pre-matched in advance based on the prediction of the electric vehicle state. Therefore, before step 202, the following steps are included:

[0073] Step 202-1a, collecting historical data of the opening degree of the electric vehicle acceleration control state, the electric vehicle motor speed and the electric vehicle acceleration as training data;

[0074] Step 202-1b, training a motor speed prediction model according to the training data;

[0075] Among them, the motor speed prediction model can be adopted in various ways, as follows:

[0076] (1) LSTM model training:

[0077] Data preparation: organize historical data in time series, and each time sampling point contains the opening degree of the electric vehicle acceleration control state, the electric vehicle acceleration, and the electric vehicle motor speed. Divide the data set into a training set and a test set; Feature engineering: preprocessing operations such as feature scaling and normalization can be performed to improve the model training effect; Build LSTM model: design the LSTM network structure, including the input layer, hidden layer, and output layer. You can try different numbers of layers and nodes to optimize the model; Model training: use the training set to train the LSTM model, and update the network parameters through the back propagation algorithm so that the model can minimize the loss function.

[0078] (2)ARIMA / SARIMA model:

[0079] Data preparation: Sort historical data into time series. Sort historical data into time series. Each time sampling point contains the opening degree of the electric vehicle acceleration control state, the electric vehicle acceleration and the electric vehicle motor speed; Determine model parameters: Use the autocorrelation function (ACF) and partial autocorrelation function (PACF) to determine the p, d and q parameters of the ARIMA / SARIMA model. If the data has seasonality (mainly for SARIMAX, the SARIMAX model is a powerful time series analysis and forecasting model. It is an extension of the ARIMA model, adding consideration of seasonal components and external factors), it is also necessary to determine the seasonal parameters; Fit model: Fit the ARIMA / SARIMA model using the determined parameters; Model training: Train the ARIMA / SARIMA model using historical data. The ARIMA model consists of three main parts: the autoregressive (AR) part, the difference (I) part and the moving average (MA) part. The formula is as follows:

[0080] Y t =c+φ1Y t-1 +φ2Y t-2 +...+φ p Y t-p +θ1e t-1 +θ2e t-2 +...+θ q et-q +e t

[0081] Y t is the observed value of the time series, i.e., the motor speed. c is a constant term. φ1,φ2,...,φ p is the autoregressive coefficient, which represents the linear combination weight of the observations at the current moment and the observations at the past p moments. θ1,θ2,...,θ q is the moving average coefficient, which represents the linear combination weight of the white noise error at the current moment and the past q moments. t is the white noise error term.

[0082] The SARIMA model is a seasonal extension of the ARIMA model, and the formula is as follows:

[0083] Y t =c+φ1Y t-1 +φ2Y t-2 +...+φ p Y t-p +θ1e t-1 +θ2e t-2 +...+θ q e t-q +Φ1Y t-s +Φ2Y t-2s +...+Φ P Y t-Ps +e t

[0084] Y t is the observed value of the time series. c is a constant term. φ1,φ2,...,φ p is the autoregressive coefficient, which represents the linear combination weight of the observations at the current moment and the observations at the past p moments. θ1,θ2,...,θ q is the moving average coefficient, which represents the linear combination weight of the white noise error at the current moment and the past q moments. Φ1,Φ2,...,Φ P is the seasonal autoregressive coefficient, which represents the linear combination weight of the observations at the current moment and the past P seasonal cycles. s is the length of the seasonal cycle. e t is the white noise error term.

[0085] The seasonality of the data is taken into account on the basis of the ARIMA model. When determining the parameters of the ARIMA / SARIMA model, the autocorrelation function (ACF) and partial autocorrelation function (PACF) are usually used to help determine the order p, d and q of the model. The SARIMA model adds consideration of seasonality on the basis of the ARIMA model. The order of the seasonal part includes the seasonal autoregressive part (SAR part), the seasonal difference part (SI part) and the seasonal moving average part (SMA part). In the SARIMA model, in addition to p, d and q, there are P, D and Q, which represent the order of the seasonal autoregressive part, the seasonal difference part and the seasonal moving average part respectively.

[0086] (3) Holt-Winters model training process:

[0087] Data preparation: organize historical data in time series, organize historical data in time series, each time sampling point contains the opening degree of the electric vehicle acceleration control state, the electric vehicle acceleration and the electric vehicle motor speed; build model: initialize trend and seasonal parameters; model training: use historical data to train the Holt-Winters model, which is a commonly used seasonal time series prediction model used to predict data trends and seasonal changes in the future. It contains three components: trend, seasonality and cycle, and is usually used to process time series data with obvious seasonal characteristics. The formula of the Holt-Winters model can be described as:

[0088]

[0089] in, is the predicted value of the motor speed at time t, L t is the level value at time t, T t The basic trend of the time series is the trend value at time t, which indicates the growth or decline trend of the time series. t-m+(h-1)modm is the seasonal value at time t, indicating periodic fluctuations, where m represents the length of the seasonal cycle and h is the predicted time step.

[0090] Step 202-1c: predict the acceleration of the electric vehicle and the motor speed of the electric vehicle according to the motor speed prediction model.

[0091] In the fourth embodiment of the present invention, the acceleration of the electric vehicle and the motor speed of the electric vehicle are predicted according to the motor speed prediction model, and the engine audio is pre-matched and read in advance, thereby avoiding the technical problem of discontinuous sound caused by processing delay.

[0092] Embodiment 5

[0093] Embodiments 1 to 4 of the present invention are mainly based on the original collected engine audio clips for subsequent sound effect synthesis, but cannot support more types or styles of throttle sound effects, and collecting more types or styles of throttle sound effects will result in higher investment costs. In order to achieve richer and more diverse sound effects and reduce the cost of sound collection, this embodiment uses the relevant technology of artificial intelligence content generation AIGC to achieve diversified throttle sound effect synthesis through audio generation methods. The steps are as follows:

[0094] In the case of active engine sound simulation, the original captured segmented engine audio can be used to generate, modify or enhance the sound of the engine. Here are the specific steps on how to apply this method:

[0095] Step 501: Prepare engine sound samples under different working conditions, including sounds under various speeds, loads and operating conditions.

[0096] Step 502: pre-process the recorded sound, such as noise reduction, normalization, etc., to facilitate model processing.

[0097] Step 503: Perform spectrum analysis on the collected sound samples to extract audio features, such as spectrogram, Mel spectrum, etc.

[0098] Step 504: Use the extracted audio features as input to the model, or directly use the original audio data as input to the model.

[0099] Step 505: Based on the audio segment synthesis method, the engine sound samples are trained through the variational autoencoder VAE to obtain a throttle sound effect generation model.

[0100] Among them, the variational autoencoder VAE is an unsupervised learning algorithm used to learn the distribution of latent variables in the data. In music generation, VAE can model different styles, emotions, and creativity by learning the latent space. By mapping the original audio data to a low-dimensional latent space and sampling it through reparameterization techniques, VAE can generate audio samples with diversity and creativity. During the training process, the model will learn the potential representation of the engine sound, including the sound characteristics under different working conditions.

[0101] The throttle sound effect generation model can be continuously adjusted and optimized during subsequent use, for example: Sound generation and adjustment: Using the trained model, new sound samples are generated by sampling or interpolation in the latent space. According to the working conditions to be simulated, the parameters in the latent space are adjusted to produce specific sound characteristics, such as higher pitch, lower noise level, or specific sound texture; Sound enhancement and optimization: If the goal is to enhance or improve the sound of an existing engine, the original sound can be superimposed or mixed with the sound generated by the model to achieve the desired sound effect (for example, synthesizing sound effects that conform to the styles of anime, two-dimensional, technology, punk, etc.). The final sound output can be further optimized by adjusting the mixing ratio or applying filters.

[0102] Further, as Figure 2 The specific implementation of the method, the embodiment of the present invention provides an electric vehicle throttle sound effect linkage generation method device, such as Figure 4 As shown, the device comprises:

[0103] The acquisition module 410 is used to acquire the electric vehicle status information, wherein the electric vehicle status information includes the electric vehicle acceleration control state, the electric vehicle motor speed and the electric vehicle acceleration;

[0104] A query module 420, configured to obtain matching engine audio according to the electric vehicle state information;

[0105] A synthesis module 430, for synthesizing a simulated throttle sound effect according to the matched engine audio;

[0106] The output module 440 is used to control the power amplifier to drive the speaker to output the simulated throttle sound effect according to the simulated throttle sound effect.

[0107] In an embodiment of the present invention, a computer device is provided, including a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, the steps of implementing the method for generating a linkage sound effect of an electric vehicle throttle include:

[0108] Acquiring electric vehicle status information, wherein the electric vehicle status information includes electric vehicle acceleration control state, electric vehicle motor speed and electric vehicle acceleration;

[0109] According to the electric vehicle state information, obtaining matching engine audio;

[0110] Synthesize simulated throttle sound effects based on matching engine audio;

[0111] According to the simulated throttle sound effect, the power amplifier is controlled to drive the speaker to output the simulated throttle sound effect.

[0112] In an embodiment of the present invention, a computer-readable storage medium is provided, on which a computer program is stored. When the computer program is executed by a processor, the following steps are implemented:

[0113] Acquiring electric vehicle status information, wherein the electric vehicle status information includes electric vehicle acceleration control state, electric vehicle motor speed and electric vehicle acceleration;

[0114] According to the electric vehicle state information, obtaining matching engine audio;

[0115] Synthesize simulated throttle sound effects based on matching engine audio;

[0116] According to the simulated throttle sound effect, the power amplifier is controlled to drive the speaker to output the simulated throttle sound effect.

[0117] It should be noted that the above embodiments only use electric bicycles to illustrate the principles and implementation steps of the embodiments of the present invention, and no specific limitation is made to the actual means of transportation. For example, the technical solution of the present invention can also be applied to electric bicycles, motorcycles, cars, etc. For the functions or steps that can be implemented by computer-readable storage media or computer devices, please refer to the aforementioned method embodiments. To avoid repetition, they will not be described one by one here.

[0118] Those skilled in the art can understand that all or part of the processes in the above-mentioned embodiment methods can be completed by instructing the relevant hardware through a computer program, and the computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to memory, storage, database or other media used in the embodiments provided by the present invention can include non-volatile and / or volatile memory. Non-volatile memory can include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM) or flash memory. Volatile memory can include random access memory (RAM) or external cache memory. As an illustration and not limitation, RAM is available in many forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous link (Synchlink) DRAM (SLDRAM), memory bus (Rambus) direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM).

[0119] Those skilled in the art can clearly understand that for the convenience and simplicity of description, only the division of the above-mentioned functional units and modules is used as an example. In actual applications, the above-mentioned functions can be distributed and completed by different functional units and modules as needed, that is, the internal structure of the device can be divided into different functional units or modules to complete all or part of the functions described above.

[0120] The embodiments described above are only used to illustrate the technical solutions of the present invention, rather than to limit the same. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that the technical solutions described in the aforementioned embodiments may still be modified, or some of the technical features may be replaced by equivalents. Such modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present invention, and should all be included in the protection scope of the present invention.

Claims

1. A method for generating a linkage sound effect of an electric vehicle throttle, characterized in that: include: Acquiring electric vehicle status information, wherein the electric vehicle status information includes electric vehicle acceleration control state, electric vehicle motor speed and electric vehicle acceleration; Collecting audio of the sound-generating component to be simulated in different working states, wherein the working states include an idle state, an acceleration state, and a deceleration state; The collected audio is processed in sections according to different preset speed ranges; Perform short-time Fourier transform on the segmented audio to obtain the frequency spectrum distribution corresponding to the segmented audio; According to the frequency spectrum distribution, the segmented audio is filtered to extract frequency components with energy accounting for more than 80%, and generate audio samples; The audio samples are stored correspondingly to the rotation speed of the fuel engine to be simulated, and a segmented audio sample library is generated; When it is determined according to the electric vehicle state information that the current acceleration is greater than zero and the electric vehicle acceleration control state is in a failed state, it is determined that the engine is in an idle state, and according to the electric vehicle motor speed, an engine audio sequence matching the idle state is obtained from the segmented audio sample library; When it is determined according to the electric vehicle state information that the current acceleration is less than zero and the electric vehicle acceleration control state is a braking state, it is determined that the engine is in a deceleration state, and based on the electric vehicle motor speed, an engine audio sequence matching the deceleration state is obtained from the segmented audio sample library; When it is determined according to the electric vehicle state information that the current acceleration is greater than zero and the electric vehicle acceleration control state is the acceleration state, it is determined that the engine is in the acceleration state, and based on the electric vehicle motor speed, an engine audio sequence matching the acceleration state is obtained from the segmented audio sample library; Synthesize simulated throttle sound effects based on matching engine audio; According to the simulated throttle sound effect, the power amplifier is controlled to drive the speaker to output the simulated throttle sound effect.

2. The electric vehicle throttle sound effect linkage generation method according to claim 1, characterized in that: The step of synthesizing the simulated throttle sound effect according to the matched engine audio comprises: The engine audio sequence obtained from the segmented audio sample library is spliced ​​according to a preset overlapping time interval to generate a spliced ​​audio; Determine the audio gain coefficient according to the opening degree of the acceleration control state of the electric vehicle; The amplitude of the spliced ​​audio is adjusted according to the audio gain factor to synthesize a simulated switch sound effect.

3. The electric vehicle throttle sound effect linkage generation method according to claim 1, characterized in that: The step of collecting audio of the sound-emitting component to be simulated in different working states specifically includes: The audio signals of the fuel engine to be simulated in different working states are collected from inside and outside the vehicle respectively; The step of acquiring the matching engine audio according to the electric vehicle status information specifically includes acquiring the matching in-vehicle engine audio and out-vehicle engine audio respectively according to the electric vehicle status information.

4. The electric vehicle throttle sound effect linkage generation method according to claim 1, characterized in that: The step of storing the audio samples corresponding to the engine speed to generate a segmented audio sample library comprises: Obtain the speed range of the electric vehicle motor and the speed range of the fuel engine to be simulated; According to the proportionality coefficient between the speed range of the electric vehicle motor and the speed range of the fuel engine to be simulated, the speed of the fuel engine to be simulated corresponding to the audio sample is converted into the speed of the electric vehicle motor; The audio samples and the converted electric vehicle motor speeds are stored correspondingly to generate a segmented audio sample library.

5. The electric vehicle throttle sound effect linkage generation method according to claim 1, characterized in that: Before the step of acquiring the matching engine audio according to the electric vehicle state information, the method includes: Collecting historical data of the opening degree of the electric vehicle acceleration control state, the electric vehicle motor speed, and the electric vehicle acceleration as training data; According to the training data, train the motor speed prediction model; According to the motor speed prediction model, the acceleration of the electric vehicle and the motor speed of the electric vehicle are predicted.

6. A method and device for generating linkage sound effects of electric vehicle throttles, characterized in that: include: An acquisition module, used to acquire electric vehicle status information, wherein the electric vehicle status information includes electric vehicle acceleration control state, electric vehicle motor speed and electric vehicle acceleration; A query module is used to collect audio of the sound-generating component to be simulated in different working states, wherein the working states include idling state, acceleration state and deceleration state; segmenting the collected audio according to different preset speed ranges; performing short-time Fourier transform on the segmented audio to obtain the spectrum distribution corresponding to the segmented audio; filtering the segmented audio according to the spectrum distribution, extracting the frequency components with energy accounting for more than 80%, and generating audio samples; storing the audio samples in correspondence with the speed of the fuel engine to be simulated, and generating a segmented audio sample library; judging that the current acceleration is greater than zero and the acceleration control state of the electric vehicle is in a failed state according to the electric vehicle state information When the engine is judged to be in an idle state, and according to the electric vehicle motor speed, an engine audio sequence matching the idle state is obtained from the segmented audio sample library; when it is judged according to the electric vehicle state information that the current acceleration is less than zero and the electric vehicle acceleration control state is a braking state, the engine is judged to be in a deceleration state, and according to the electric vehicle motor speed, an engine audio sequence matching the deceleration state is obtained from the segmented audio sample library; when it is judged according to the electric vehicle state information that the current acceleration is greater than zero and the electric vehicle acceleration control state is an acceleration state, the engine is judged to be in an acceleration state, and according to the electric vehicle motor speed, an engine audio sequence matching the acceleration state is obtained from the segmented audio sample library; Synthesis module, used to synthesize simulated throttle sound effects based on matching engine audio; The output module is used to control the power amplifier to drive the speaker to output the simulated throttle sound effect according to the simulated throttle sound effect.

7. A computer device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that: When the processor executes the computer program, the steps of the method for generating the linkage sound effect of the electric vehicle throttle as described in any one of claims 1 to 5 are implemented.

8. A computer-readable storage medium storing a computer program, characterized in that: When the computer program is executed by a processor, the steps of the method for generating linkage sound effects of electric vehicle throttles as described in any one of claims 1 to 5 are implemented.

Images