Multi-channel audio file generation method and device, computer equipment and storage medium

Abstract

The invention discloses a multichannel audio file generation method and device, computer equipment and a storage medium, and the method comprises the steps: carrying out the human voice separation of a stereo audio file, and obtaining a human voice signal and an accompaniment sound signal; determining the arrangement positions of a plurality of loudspeakers in the target sound system, and determining the sound channel gain of each loudspeaker according to the arrangement positions of the plurality of loudspeakers; and generating audio signals corresponding to the loudspeakers based on the human voice signals, the accompaniment sound signals and the sound channel gains of the loudspeakers, and combining the audio signals into a multi-sound-channel audio file adaptive to the target sound system. The purpose of converting the stereo audio file into different multi-channel formats to adapt to different multi-channel sound systems is achieved, and a user can determine the channel gain of each loudspeaker according to the number and arrangement positions of the loudspeakers of the target sound system. Therefore, the stereo audio file is converted into the multi-channel audio file adaptive to the target sound system, and the use experience of a user on the multi-channel sound system can be improved.

Description

Technical Field

[0001] The present invention relates to the field of audio processing technology, and in particular to a method, device, computer equipment and storage medium for generating a multi-channel audio file. Background Art

[0002] With the development of society, people's auditory requirements for audio are also increasing. The traditional stereo sound system with left and right speakers can no longer meet the needs of some users. Therefore, more immersive and realistic multi-channel sound systems have come into being, such as 5.1.2 speaker, 7.1 speaker, and 7.1.4 speaker sound systems.

[0003] Multi-channel audio systems with different speaker configurations require different audio formats. Users often need to find suitable audio signals to experience the effects of the multi-channel audio system. However, music and other audio on the market are currently mainly stereo audio files, and the number of audio sources adapted to the multi-channel audio system is relatively small. For example, many music songs do not have multi-speaker formats. Users have fewer audio signals to choose from when using the multi-channel audio system, resulting in a poor user experience of the multi-channel audio system. Summary of the invention

[0004] The present invention provides a multi-channel audio file generation method, device, computer equipment and storage medium, which converts stereo audio files into audio files in different multi-channel formats so as to adapt to different multi-channel sound systems, so as to solve the problem that the number of sound sources adapted to the multi-channel sound system is small, resulting in a poor user experience of the multi-channel sound system.

[0005] In view of the above problems, a method for generating a multi-channel audio file is provided, comprising:

[0006] Separate the input stereo audio file into human voice to obtain human voice signal and accompaniment sound signal;

[0007] Determine the arrangement positions of multiple speakers in the target sound system, and determine the channel gain of each speaker according to the arrangement positions of the multiple speakers;

[0008] The audio signals of the corresponding speakers are generated based on the vocal signal, the accompaniment sound signal and the channel gain of the speakers, and are merged into a multi-channel audio file adapted to the target sound system.

[0009] A multi-channel audio file generating device is provided, comprising:

[0010] The separation module is used to separate the human voice from the input stereo audio file to obtain a human voice signal and an accompaniment sound signal;

[0011] A determination module, used to determine the arrangement positions of multiple speakers in the target sound system, and determine the channel gain of each speaker according to the arrangement positions of the multiple speakers;

[0012] The generation module is used to generate audio signals corresponding to the speakers based on the human voice signal, the accompaniment sound signal and the channel gain of the speakers, and merge them into a multi-channel audio file adapted to the target sound system.

[0013] A computer device is provided, 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 multi-channel audio file generating method when executing the computer program.

[0014] A computer-readable storage medium is provided, wherein the computer-readable storage medium stores a computer program, and wherein the computer program implements the steps of the above-mentioned multi-channel audio file generation method when executed by a processor.

[0015] In a solution provided by the above-mentioned multi-channel audio file generation method, device, computer equipment and storage medium, the input stereo audio file is separated from the human voice to obtain the human voice signal and the accompaniment sound signal, and then the layout positions of multiple speakers in the target audio system are determined, and the channel gain of each speaker is determined according to the layout positions of the multiple speakers, and then the audio signal of the corresponding speaker is generated based on the human voice signal, the accompaniment sound signal and the channel gain of the speaker, and merged into a multi-channel audio file adapted to the target audio system; it realizes the purpose of converting the stereo audio file into different channel formats to adapt to different multi-channel audio systems, solves the problem of less sound sources adapted to the multi-channel audio system, and the user can determine the gain of each speaker according to the number of speakers in the target audio system and their layout positions, so as to flexibly convert the stereo audio file into a multi-channel audio file adapted to the target audio system, which can improve the user experience of the multi-channel audio system. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings required for use in the description of the embodiments of the present invention will be briefly introduced below. Obviously, the accompanying drawings in the following description are only some embodiments of the present invention. For ordinary technicians in this field, other accompanying drawings can be obtained based on these accompanying drawings without paying creative labor.

[0017] Figure 1 is a structural schematic diagram of a multi-channel audio file generation system in one embodiment of the present invention;

[0018] Figure 2 is a flow chart of a method for generating a multi-channel audio file in one embodiment of the present invention;

[0019] Figure 3 is an audio spectrum of a stereo audio file in one embodiment of the present invention;

[0020] Figure 4 Yes Figure 3 The audio spectrum of the 7.1-channel audio file obtained by converting the stereo audio file;

[0021] Figure 5 yes Figure 2 A schematic diagram of an implementation process of step S20;

[0022] Figure 6 , Figure 7 This is a schematic diagram of the placement of the speakers in a 7.1.4-channel sound system;

[0023] Figure 8 is a structural schematic diagram of a multi-channel audio file generating device in one embodiment of the present invention;

[0024] Fig. 9 It is another structural schematic diagram of a multi-channel audio file generating device in one embodiment of the present invention. DETAILED DESCRIPTION

[0025] 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 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.

[0026] The multi-channel audio file generation method provided by the embodiment of the present invention can be applied in Figure 1 The multi-channel audio file generation system includes a multi-channel audio file generation device, a target sound system and a terminal device. The terminal device and the target sound system communicate with the multi-channel audio file generation device through a network. The target sound system is a multi-channel sound system, which includes a plurality of speakers.

[0027] When a stereo audio file needs to be converted into a format for multiple speakers, the user inputs (can be input through a terminal device) the stereo audio file to be converted into the multi-channel audio file generating device; then the multi-channel audio file generating device separates the input stereo audio file into human voices to obtain human voice signals and accompaniment sound signals; determines the arrangement positions of multiple speakers in the target audio system, and determines the channel gain of each speaker according to the arrangement positions of the multiple speakers; then generates audio signals of corresponding speakers based on the human voice signals, accompaniment sound signals and the channel gains of the speakers, and merges them into a multi-channel audio file adapted to the target audio system; finally, the multi-channel audio file is directly sent to the target audio system for playback, or exported to the terminal device for storage, so as to be subsequently transmitted to the target audio system for playback. The method proposed in this embodiment achieves the purpose of converting a stereo audio file into different multi-channel formats to adapt to different multi-channel sound systems, solves the problem that the number of sound sources adapted to the multi-channel sound system is small and it is difficult for users to find suitable sound sources. The user can determine the gain of each speaker (that is, determine the weight of each speaker) according to the number of speakers of the target sound system and their corresponding layout positions, thereby flexibly converting the stereo audio file into a multi-channel audio file adapted to the target sound system, which can improve the user experience of the multi-channel sound system.

[0028] In this embodiment, the terminal device can be, but is not limited to, various personal computers, laptops, smart phones, tablet computers, and portable wearable devices. The multi-channel audio file generation device can be various personal computer devices, and can also be implemented by an independent server or a server cluster composed of multiple servers.

[0029] In one embodiment, if Figure 2 As shown, a method for generating a multi-channel audio file is provided, and the method is applied in Figure 1 The multi-channel audio file generation device in the embodiment of the present invention is described as an example, comprising the following steps:

[0030] S10: Separate the stereo audio file into human voices to obtain a human voice signal and an accompaniment sound signal.

[0031] When a stereo audio file needs to be converted into a multi-speaker format, the user inputs the stereo audio file to be converted into the multi-channel audio file generating device, and then the multi-channel audio file generating device separates the input stereo audio file into human voices to obtain human voice signals and accompaniment sound signals.

[0032] Among them, after obtaining the stereo audio file input by the user, the multi-channel audio file generation device can obtain a pre-trained stereo separation model, and then input the stereo audio file into the stereo separation model, and separate the stereo audio file from the human voice through the stereo separation model to obtain a human voice signal and an accompaniment sound signal. In this embodiment, using a pre-trained stereo separation model to separate the human voice is simple and convenient, and the separated data is more accurate.

[0033] S20: Determine the arrangement positions of a plurality of speakers in the target sound system, and determine the channel gain of each speaker according to the arrangement positions of the plurality of speakers.

[0034] It is important to understand that different types of multi-channel audio systems have different speaker configurations (speaker configurations), that is, different multi-channel audio systems (abc channel audio systems) have different numbers of speakers and different layouts. The abc channel audio system is a system consisting of a surround speakers + b subwoofers + c sky speakers, a total of a+b+c speakers. Among them, position a represents a surround speaker; position b represents b specially designed subwoofers, which require a separate subwoofer to produce sound; position c represents c sky speakers, which are installed above the actual listening position (listener position). If c has no value, it means that no sky speakers are configured. For example, a 7.1.4 speaker system has 7 surround speakers, including center, left and right front, left and right surround, and left and right rear, plus 1 subwoofer and 4 sky speakers, for a total of 12 speakers. The number of speakers (number of speakers) in the 5.1-channel, 5.1.2-channel, 7.1-channel, and 7.1.4-channel sound systems are 6, 8, 8, and 12 respectively.

[0035] Therefore, after separating the human voice signal and the accompaniment sound signal in the stereo audio file, the multi-channel audio file generation device needs to determine the speaker configuration of the target audio system, that is, to determine the number of multiple speakers (the number of channels) and the arrangement positions of the multiple speakers (the speaker positions of the multiple channels) in the target audio system. After determining the arrangement positions of the multiple speakers in the target audio system, it is necessary to determine the channel gain of each speaker according to the arrangement positions of the multiple speakers.

[0036] S30: generating audio signals corresponding to the speakers based on the human voice signal, the accompaniment sound signal and the channel gains of the speakers, and merging them into a multi-channel audio file adapted to the target sound system.

[0037] After determining the channel gain of each speaker, the multi-channel audio file generating device also needs to generate the audio signal of the corresponding speaker based on the human voice signal, the accompaniment sound signal and the channel gain of the speaker, and merge them into a multi-channel audio file adapted to the target sound system. For example, the human voice signal, the accompaniment sound signal and the channel gain of the speaker are directly allocated to the channel of the corresponding speaker to generate the audio signal of the corresponding speaker, and then the audio signals of each speaker are merged to obtain a multi-channel audio file adapted to the target sound system.

[0038] When determining the channel gain of each speaker, the layout position of each speaker of the target sound system is taken into consideration. On the one hand, the number and layout of the speakers can be flexibly changed so that the subsequently generated multi-channel audio files can be adapted to different multi-channel sound systems; on the other hand, the audio playback effect of the corresponding speaker can be adjusted according to the channel gain of each speaker, thereby improving the playback effect of the entire target sound system and enhancing the user's auditory experience.

[0039] The method proposed in this embodiment solves the problem that the number of sound sources adapted to the multi-channel sound system is small and it is difficult for users to find suitable sound sources. The user can determine the gain of each speaker (that is, determine the weight of each speaker) according to the number of multiple speakers in the target sound system and their corresponding layout positions, thereby flexibly converting a stereo audio file into a multi-channel audio file adapted to the target sound system, which can improve the user experience of the multi-channel sound system.

[0040] In addition, currently the music tracks on the market are mainly stereo audio, while there are many types of multi-channel audio systems, such as 5.1-channel, 5.1.2-channel, 7.1-channel, and 7.1.4-channel audio systems. If the music track is to be adapted to all types of multi-channel audio systems, the stereo audio needs to be converted into audio in different multi-speaker formats for download by different users. The production cost is high and it will increase the amount of network data and storage. In addition, in order to maximize the effect, the multi-speaker audio sources on the market require that the position and number of each speaker in the multi-channel audio system are fixed. In fact, due to many factors (such as space constraints in the car), users may not be able to meet these requirements well, resulting in poor music playback.

[0041] The method provided in this embodiment can effectively solve the above problems by decomposing a stereo music track into a vocal signal and an accompaniment signal, determining the gain of each speaker according to a certain algorithm, and then reallocating the vocal signal and the accompaniment signal to the corresponding speakers according to the gain (weight) of each speaker. First, the algorithm can be flexibly changed according to the number and layout of speakers of different multi-channel audio systems, so as to adapt to different multi-channel audio systems. Users can convert multi-channel audio files that match the number of speakers according to the type of their own audio system, without converting all types of audio files, thereby reducing production costs and storage volume. Second, users can create audio multi-channel audio files suitable for their own audio system according to the number and layout of speakers of their own audio system to maximize auditory enjoyment.

[0042] Take a stereo audio file as an example. The audio spectrum of the stereo audio file is as follows: Figure 3 As shown, the method provided in this embodiment is used to generate a multi-channel audio file adapted to a 7.1-channel sound system, that is, a 7.1-channel format audio file. The audio spectrum of the 7.1-channel format audio file is as shown in FIG. Figure 4 As shown. Figure 3 and Figure 4 It can be seen from the comparison that compared with the traditional two-channel stereo audio file, the multi-channel audio file of the present application has a more immersive and enveloping listening experience, which can greatly enhance the user's auditory enjoyment.

[0043] In one embodiment, in step S10, the input stereo audio file is subjected to vocal separation to obtain a vocal signal and an accompaniment sound signal, which specifically includes the following steps:

[0044] S11: Obtain a pre-trained stereo separation model, where the stereo separation model includes a signal transformation layer and a signal filter.

[0045] After obtaining the stereo audio file input by the user, the multi-channel audio file generation device needs to obtain a pre-trained stereo separation model, so as to separate the human voice from the stereo audio file through the stereo separation model to obtain a human voice signal and an accompaniment sound signal. The stereo separation model is a neural network model, which includes a signal transformation layer and a signal filter.

[0046] S12: Input the stereo audio file into the stereo separation model, perform Fourier transform through the signal transformation layer to obtain a transformed audio signal, and perform vocal separation on the transformed audio signal through a signal filter to obtain a vocal signal and an accompaniment sound signal.

[0047] After obtaining the pre-trained stereo separation model, the stereo separation model is called, and the stereo audio file is input into the stereo separation model to implement the following steps: Fourier transform is performed through the signal transformation layer to obtain a transformed audio signal, and the transformed audio signal is separated from the human voice through the signal filter to obtain a human voice signal and an accompaniment sound signal.

[0048] In this embodiment, a stereo separation model obtained by pre-training is obtained, and then a stereo audio file is input into the stereo separation model, a Fourier transform is performed through a signal conversion layer to obtain a transformed audio signal, and a human voice signal and an accompaniment sound signal are obtained by separating the human voice from the input stereo audio file through a signal filter, and the specific steps of obtaining the human voice signal and the accompaniment sound signal are clarified. The use of a neural network model including a signal conversion layer and a signal filter to separate the human voice from a stereo audio file is simple, convenient and highly accurate; and in the process of training the stereo separation model, the network parameters of the signal conversion layer and the signal filter are iterated at the same time, which improves the coordination between the signal conversion layer and the signal filter, thereby improving the accuracy of human voice separation.

[0049] In other embodiments, after obtaining the stereo audio file input by the user, the multi-channel audio file generating device first performs Fourier transform on the stereo audio file based on the Fourier algorithm to obtain a transformed audio signal, and then inputs the transformed audio signal into a signal filter to separate the transformed audio signal from the human voice to obtain a human voice signal and an accompaniment sound signal. The human voice separation can be performed using a conventional signal filter, which can meet a certain separation effect without the need for model training.

[0050] Among them, the signal filter can also be a filter that is obtained by iteratively training the network parameters of the filter based on stereo audio (including human voice and accompaniment sound) and the standard human voice of the stereo audio, and the effect meets the requirements. The training process is: first, the stereo audio is Fourier transformed to obtain the transformed stereo audio, and then the transformed stereo audio is input into the filter for human voice separation to obtain human voice and accompaniment sound, and then the residual of human voice and accompaniment sound is compared, and then the parameters of the filter are updated according to the residual, and the human voice separation process is repeated until the residual of human voice and accompaniment sound meets the requirements, and the human voice and accompaniment that meet the effect of the filter separation are determined, and the filter converges, and then the filter of the current parameters is output as a signal filter for subsequent use. The signal filter obtained by training in this way can further improve the accuracy of the signal filter, thereby improving the accuracy of the signal data obtained by subsequent movement human voice separation.

[0051] In one embodiment, before step S10, a stereo separation model needs to be pre-trained so that after the input stereo audio file is subsequently acquired, the stereo audio file can be separated from the human voice through the stereo separation model to obtain the human voice signal and the accompaniment sound signal. Specifically, the stereo separation model is trained in the following manner:

[0052] S01: Acquire stereo audio samples including vocals and accompaniment sounds, and standard vocals of the stereo audio samples.

[0053] When training the model, first, the multi-channel audio file generation device needs to obtain stereo audio samples including human voice and accompaniment, as well as standard human voice of the stereo audio samples. The stereo audio sample can be one sample or multiple samples. Using one stereo audio sample for model training, the amount of iterative data is small, the number of iterations can be reduced, and a stereo separation model that meets the separation effect requirements can be quickly obtained; using multiple stereo audio samples for model training, the more stereo audio samples there are, the more iterative data there is, the more iterations there are, but the training accuracy of the stereo separation model can be improved, and the stereo audio samples can be determined according to actual needs.

[0054] S02: Input the stereo audio sample into the pre-trained neural network model, perform Fourier transform through the Fourier transform network to obtain the Fourier signal, and perform vocal separation on the Fourier signal through the filter to obtain the separated vocal and the separated accompaniment sound.

[0055] In this embodiment, the pre-trained neural network model includes a Fourier transform network and a filter. The multi-channel audio file generation device inputs the stereo audio sample into the pre-trained neural network model, performs Fourier transform through the Fourier transform network to obtain a Fourier signal, and performs vocal separation on the Fourier signal through the filter to obtain separated vocals and separated accompaniment sounds.

[0056] S03: determining the residual between the separated human voice and the standard human voice;

[0057] S04: Iteratively update the parameters of the filter according to the residual until the separated human voice obtained by the filter separation meets the effect requirements, and output the current Fourier transform network and filter as a stereo separation model.

[0058] After obtaining the separated vocals and the separated accompaniment sound, the multi-channel audio file generation device needs to determine the residual of the separated vocals and the standard vocals, and then iteratively update the parameters of the filter according to the residual until the separated vocals obtained by the filter separation meet the effect requirements, and then output the current Fourier transform network and filter as a stereo separation model. Specifically, after determining the residual of the separated vocals and the standard vocals, it is necessary to determine whether the residual is less than the preset residual value; if the residual is greater than the preset residual value, it is determined that the separation effect of the filter does not meet the requirements, then adjust the network parameters of the filter, and repeat steps S02-S03 until the residual is less than or equal to the preset residual value, and it is determined that the separated vocals and accompaniment sound obtained by the filter separation meet the effect requirements. At this time, it is determined that the filter converges and can separate the vocals and accompaniment sound that meet the effect requirements, then output the current Fourier transform network and filter as a stereo separation model. Among them, the output current Fourier transform network is a signal conversion layer, and the output current filter is a signal filter.

[0059] In this embodiment, a stereo audio sample including a human voice and an accompaniment sound, and a standard human voice of the stereo audio sample are obtained; the stereo audio sample is input into a pre-trained neural network model, a Fourier transform is performed through a Fourier transform network to obtain a Fourier signal, and the Fourier signal is separated from the human voice through a filter to obtain a separated human voice and a separated accompaniment sound; the residual between the separated human voice and the standard human voice is determined; the parameters of the filter are iteratively updated according to the residual until the separated human voice obtained by the filter separation meets the effect requirements, and the current Fourier transform network and filter are output as a stereo separation model; the training process of the stereo separation model is clarified, and through the above training method, a stereo separation model with better separation effect can be trained, which provides a solid foundation for the subsequent separation of human voices from stereo audio files.

[0060] In addition, in the stereo separation model training process of this embodiment, only the parameters of the filter are iteratively updated, which can reduce the number of network parameter iterations, thereby accelerating the model convergence speed. In other embodiments, during the process of training the stereo separation model, the parameters of the Fourier transform network can also be synchronously updated to further improve the accuracy of the stereo separation model.

[0061] In one embodiment, if Figure 5 As shown, in step S20, the channel gain of each speaker is determined according to the arrangement positions of the multiple speakers, which specifically includes the following steps:

[0062] S21: setting an actual listening position in the target sound system, and determining the listening distance of each speaker according to the actual listening position and the arrangement position of the speakers.

[0063] S22: Determine the channel gain of each speaker according to the listening distance of the speaker.

[0064] After determining the arrangement positions of the multiple speakers of the target sound system, it is necessary to set an actual listening position (i.e., the listener position) in the target sound system. In this embodiment, the actual listening position in the target sound system may not coincide with the best listening position of the target sound system, or may coincide with the best listening position of the target sound system.

[0065] It is understandable that the best listening position of the target sound system is located at the center of the symmetrical figure formed by the surround speakers. For a multi-channel sound system, the placement of the surround speakers is generally based on the placement of the center speaker. Specifically, an xyz coordinate system is first established, the xy plane is parallel to the horizontal plane, the coordinate origin points to the placement direction of the center speaker as the positive direction of the y axis, the z axis is perpendicular to the xy plane, and the other speakers are arranged around the coordinate origin at different heights. In the multi-channel sound system, the surround speakers except the center speaker are symmetrical relative to the yz plane, and the sky speakers are symmetrical relative to the yz plane, forming multiple symmetrical speaker groups. At this time, the coordinate origin is the center of the multi-channel sound system, that is, the best listening position. According to the number of speakers in different multi-channel sound systems, the multiple symmetrical speaker groups include 2 or 3 symmetrical speaker groups corresponding to the surround speakers: a main symmetrical speaker group, a surround symmetrical speaker group, a rear symmetrical speaker group (some multi-channel sound systems do not have the rear symmetrical speaker group), and 1 or 2 symmetrical speaker groups corresponding to the sky speakers. Taking the 7.1.4 channel sound system as an example, the placement of the speakers is as follows: Figure 6 , Figure 7 As shown in the figure, the 7.1.4 channel sound system includes 7 surround speakers: right speaker R, center speaker C, left speaker L, right surround speaker Rs, left surround speaker Ls, left rear speaker Lb and right rear speaker Rb; 1 subwoofer LEF and 4 sky speakers: upper left front surround speaker Ltf, upper right front surround speaker Rtf, upper left rear surround speaker Ltb and upper right rear surround speaker Rtb. Among them, O is the origin of coordinates (i.e. the best listening position), the direction from the origin of coordinates O to the placement position of the center speaker C is the positive direction of the y-axis, and the 5 speaker groups of R and L, Rs and Ls, Lb and Rb, Ltf and Rtf, and Ltb and Rtb are symmetrical speaker groups that are symmetrical in pairs relative to the yz plane. R and L, Rs and Ls, and Lb and Rb are 3 symmetrical speaker groups corresponding to the surround speakers; Ltf and Rtf, and Ltb and Rtb are 2 symmetrical speaker groups corresponding to the sky speakers.

[0066] After setting an actual listening position in the target sound system, it is necessary to determine the listening distance of each speaker according to the actual listening position and the arrangement position of the speaker. The listening distance of the speaker is the straight-line distance between the speaker and the actual listening position; then determine the channel gain of each speaker according to the straight-line distance between each speaker and the actual listening position, that is, determine the channel gain of the speaker according to the listening distance of the speaker, so as to obtain the channel gain of each speaker. For example, the farther the straight-line distance between the speaker and the actual listening position, the greater the channel gain corresponding to the speaker, and the closer the straight-line distance between the speaker and the actual listening position, the smaller the channel gain corresponding to the speaker, so as to ensure that the same audio signal can be heard at different listening positions.

[0067] When using a multi-channel sound system, different positions have different listening effects, and the best listening position has the best listening effect. Therefore, in the traditional multi-channel audio file generation method, the best listening position is usually used as a fixed standard for audio conversion, that is, the actual listening position is assumed to be the best listening position. However, in real life, due to factors such as space limitations, the audience may not be able to listen to the audio at the best listening position (for example, in complex and small space environments such as vehicles and homes), and the audience position will change in different scenes. If the audio conversion is performed with the best listening position as a fixed standard, the listening effect after the subsequent audio file is played cannot be optimal. In this embodiment, the linear distance between the speaker and the actual listening position is determined by the actual listening position and the arrangement position of the speaker, and then the channel gain of each speaker is determined. When calculating the channel gain of the speaker, the relative position of the actual listening position and each speaker is taken into account, so that the channel gain of each speaker can change with the change of the actual listening position (listener position), so that different positions within the speaker enclosure can also obtain the best listening effect, even if the listener is not in the center position, the same immersive listening experience can be obtained.

[0068] In this embodiment, an actual listening position is set in the target sound system, and the straight-line distance between the speaker and the actual listening position is determined according to the actual listening position and the arrangement position of the speaker, and then the channel gain of each speaker is determined according to the listening distance of each speaker, and the step of determining the channel gain of each speaker according to the arrangement position of multiple speakers is refined. When calculating the channel gain of the speaker, the relative position of the actual listening position and each speaker is taken into consideration, and the channel gain of each speaker can be adjusted according to the actual listening position, so that different positions within the enclosure of the speaker can also obtain the best listening effect, thereby improving the user experience.

[0069] In other embodiments, the actual listening position may be ignored, and the best listening position of the target sound system may be used as a standard to determine the straight-line distance between the speaker and the best listening position, and then determine the channel gain of each speaker. For the specific determination process, refer to the calculation process of determining the channel gain of each speaker based on the actual listening position, which will not be repeated here.

[0070] In one embodiment, the channel gain includes a vocal gain and an accompaniment gain. In step S22, the channel gain of each speaker is determined according to the listening distance of each speaker, which specifically includes the following steps:

[0071] S221: Determine the sound field width coefficient of each speaker, where the sound field width coefficient includes a vocal width coefficient and an accompaniment width coefficient.

[0072] S222: Determine the vocal gain of each speaker according to the vocal width coefficient of each speaker and the listening distance.

[0073] S223: Determine the accompaniment gain of each speaker according to the accompaniment width coefficient of each speaker and the listening distance.

[0074] After determining the listening distance of each speaker according to the actual listening position and the arrangement position of the speaker, it is also necessary to determine the sound field width coefficient of each speaker. In this embodiment, the channel gain includes the vocal gain and the accompaniment gain. Correspondingly, the sound field width coefficient includes the vocal width coefficient and the accompaniment width coefficient.

[0075] After determining the sound field width coefficient of the speaker, the vocal gain of each speaker is determined according to the vocal width coefficient of each speaker and the listening distance of each speaker (the straight-line distance between each speaker and the actual listening position). At the same time, the accompaniment gain of each speaker is determined according to the accompaniment width coefficient of each speaker and the listening distance of each speaker. The sound field width coefficient is a preset constant vector. The sound field width coefficient can control the energy contribution of each channel.

[0076] It should be understood that the sound field width of the sound signal is different, giving people different subjective hearing feelings. Different auditory effects can be achieved by controlling the sound field width. If the sound field width is too wide, it will give people a subjective hearing feeling of a whole piece of sound without a clear positioning, and the sound will sound vague and illusory, but the narrower the sound field width, the more obvious the positioning effect of the sound source, and the clearer the human voice. However, the sound field width cannot be too narrow. Among them, due to the singularity of the human voice in the audio, the sound field width of the human voice is a smaller range. For example, when a singer sings at an actual concert, the feeling of positioning is clear, and the specific sound position can be perceived but not precisely to the point. The sound field width of the human voice is narrow; on the contrary, there are many kinds of musical instruments in an actual concert, thus forming an accompaniment sound, and the various instruments are scattered. If the sound field width of the accompaniment sound is narrow, it is easy to cause different instruments to overlap in a certain position, and the actual listening experience is very poor. Only when the sound field width is wide enough can different instruments be better separated to obtain better auditory effects.

[0077] In summary, the vocal width coefficient in the sound field width coefficient needs to be different from the accompaniment width coefficient, so that the vocal width coefficient can be used to flexibly control the sound field width formed by the vocal signal, and the accompaniment width coefficient can be used to flexibly control the sound field width formed by the accompaniment signal, so as to avoid widening the vocal signal in the music and widen the accompaniment signal, thereby improving the listening effect, while enhancing the sense of immersion and surround, and restoring the listening effect of a real stage or concert to a greater extent.

[0078] In this embodiment, the sound field width coefficient can be a set of fixed constant coefficients determined in advance according to the audio playback effect test, so as to facilitate subsequent calculations. In other embodiments, multiple sound field width coefficients can also be pre-set, and each sound field width coefficient includes a corresponding vocal width coefficient and an accompaniment width coefficient, so that the multi-channel audio files generated subsequently have different hearing effects, that is, different sound field width coefficients can correspond to different hearing effects. Since different users have different hearing preferences, when generating a multi-channel audio file, the user can select a sound field width coefficient that meets his or her preferences from a plurality of sound field width coefficients according to actual needs. The sound field width coefficient includes a vocal width coefficient and an accompaniment width coefficient that meet his or her hearing preferences, so that the multi-channel audio file generation device generates the channel gain of each speaker according to the sound field width coefficient selected by the user, and then obtains a multi-channel audio file that meets the user's hearing preferences to meet the hearing needs of different users.

[0079] In this embodiment, by determining the sound field width coefficient of the speaker, the sound field width coefficient includes the vocal width coefficient and the accompaniment width coefficient, and then determining the vocal gain of each speaker according to the vocal width coefficient and the listening distance of each speaker, and determining the accompaniment gain of each speaker according to the accompaniment width coefficient and the listening distance of each speaker, the specific steps of determining the channel gain of each speaker according to the listening distance of each speaker are refined. When calculating the channel gain of each speaker, the sound field width formed by the vocal signal is flexibly controlled by the vocal width coefficient, and the sound field width formed by the accompaniment signal is flexibly controlled by the accompaniment width coefficient, so as to avoid widening the vocal signal in the music and widen the accompaniment signal, thereby improving the listening effect, while enhancing the sense of immersion and surround, restoring the listening effect of a real stage or concert to a greater extent.

[0080] In one embodiment, step S222, that is, determining the vocal gain of each speaker according to the vocal width coefficient of each speaker and the listening distance, specifically includes the following steps:

[0081] S2221: When the target sound system has a sky speaker, divide the plurality of speakers into a first speaker group and a second speaker group.

[0082] After determining whether the target sound system has a sky speaker, when it is determined that the target sound system has a sky speaker, the multiple speakers need to be divided into a first speaker group and a second speaker group according to the arrangement positions of the multiple speakers. The second speaker group consists of the sky speakers. The first speaker group may include other speakers except the sky speakers; the first speaker group may also include only surround speakers, that is, not including a woofer.

[0083] To ensure the accuracy of the channel gain of each speaker, it is necessary to consider the impact of the height position of the ceiling speaker on the channel gain, group the positions of different speakers, and use different algorithms to calculate the gains of the surround speakers and ceiling speakers.

[0084] S2222: For the first speaker group, determine the vocal gain of each speaker based on the first vocal constraint model, the vocal width coefficient of each speaker, and the listening distance.

[0085] After dividing multiple speakers into a first speaker group and a second speaker group, for the first speaker group, the vocal gain of each speaker is determined based on the first vocal constraint model, the vocal width coefficient of each speaker, and the straight-line distance between each speaker and the actual listening position (i.e., the listening distance).

[0086] The first human voice constraint model includes multiple constraint formulas. For example, it may include a first constraint formula and a second constraint formula. The first constraint formula may constrain each speaker based on the vector superposition principle of the sound signal; the second constraint formula may constrain any group of left-right symmetrical speakers based on the sound symmetry principle on the left and right sides of the listener; the second constraint formula may constrain the speakers located on the same side of the actual listening position based on the same sound effect on the same side of the listener.

[0087] S2223: For the second speaker group, determine a vocal gain of each speaker based on the second vocal constraint model and the listening distance of each speaker.

[0088] At the same time, for the second speaker group, the vocal gain of each speaker is determined based on the second vocal constraint model and the straight-line distance between each speaker and the actual listening position (i.e., the listening distance). The constraint formula of the second vocal constraint model is different from that of the first vocal constraint model. The constraint formula of the second vocal constraint model takes into account the influence of the height of the sky speaker on the listening effect.

[0089] S2224: When there is no overhead speaker in the target sound system, determine the vocal gain of each speaker based on the first vocal constraint model, the vocal width coefficient of the speaker, and the listening distance.

[0090] After determining the sound field width coefficient of the speaker, determine whether there is a sky speaker in the target sound system. When it is determined that there is no sky speaker in the target sound system, there is no need to consider the differences in sky speakers. At this time, the vocal gain of each speaker can be directly determined based on the first vocal constraint model, the vocal width coefficient of each speaker, and the straight-line distance between each speaker and the actual listening position (i.e., the listening distance).

[0091] In other embodiments, it is not necessary to determine whether there are overhead speakers in the target sound system, and the vocal gain of each speaker can be determined directly based on the first vocal constraint model, the vocal width coefficient of the speaker, and the straight-line distance between each speaker and the actual listening position, thereby reducing the number of judgment steps.

[0092] In this embodiment, when there is no sky speaker in the target sound system, the vocal gain of each speaker is determined based on the first vocal constraint model, the vocal width coefficient of the speaker and the listening distance; when there is a sky speaker in the target sound system, the multiple speakers are divided into a first speaker group and a second speaker group, and the vocal gain of each speaker in the first speaker group is determined based on the first vocal constraint model, the vocal width coefficient of each speaker and the listening distance; the vocal gain of each speaker in the second speaker group is determined based on the second vocal constraint model and the listening distance of each speaker; the specific calculation process of determining the vocal gain of each speaker according to the vocal width coefficient and the listening distance of the speaker is clarified, and different channel gain calculation methods are adopted according to the speaker configuration of different target sound systems to improve the accuracy of the channel gain technology. When there is a sky speaker in the target sound system, the influence of the height of the sky speaker on the channel gain is taken into account, so that a more accurate channel gain can be calculated.

[0093] In one embodiment, when it is determined that there is no overhead speaker in the target sound system, for example, the target sound system is a 5.1-channel sound system or a 7.1-channel sound system, the subwoofer of the target sound system can be determined, and then the preset gain is used as the vocal gain of the subwoofer. The vocal gain of each remaining speaker only needs to be determined according to the vocal width coefficient of the speaker and the straight-line distance between the speaker and the actual listening position.

[0094] In this embodiment, the preset gain may be a preset fixed gain value, for example, the preset gain may be 0. The playback device of the woofer is a subwoofer, and the purpose of the woofer is to restore low-frequency music such as drums and bass in the music as much as possible, and will not play human voices. In step S10, the human voice and accompaniment of the music have been separated, and the effect of the low-frequency accompaniment sound after gain is not much different. Therefore, a fixed gain value can be directly set as the human voice gain or accompaniment gain of the woofer to reduce the amount of data processing in the channel gain calculation process.

[0095] In one embodiment, in step S2222 or step S2224, determining the vocal gain of each speaker based on the first vocal constraint model, the vocal width coefficient of each speaker, and the listening distance specifically includes the following steps:

[0096] S201: Determine a first angle of each speaker.

[0097] When calculating the channel gain of each speaker, you first need to determine the first angle of each speaker based on the actual listening position and the placement of each speaker. The first angle is the angle between the speaker ray and the positive direction of the y-axis in the xyz coordinate system; the speaker ray is the ray formed by the actual listening position pointing to the placement of the speaker; the positive direction of the y-axis is the direction from the origin of the xyz coordinate system to the placement of the center speaker. Taking the 7.1.4-channel sound system as an example, the positive direction of the y-axis can be found in the following figure: Figure 6 .

[0098] S202: Determine a second angle of each speaker.

[0099] At the same time, it is also necessary to determine the second angle of each speaker based on the actual listening position and the placement position of each speaker. The second angle is the horizontal angle of the speaker ray, that is, the angle formed by the ray formed by the actual listening position pointing to the placement position of the speaker and the horizontal plane.

[0100] S203: Inputting the vocal width coefficient, the first angle, the second angle and the listening distance of each loudspeaker into the first vocal constraint model for solving, thereby obtaining the vocal gain of each loudspeaker.

[0101] After determining the first angle and the second angle of each speaker, the vocal width coefficient, the first angle, the second angle and the listening distance of each speaker are input into the first vocal constraint model for solution to obtain the vocal gain of each speaker.

[0102] In this embodiment, by determining the first angle and the second angle of each speaker, wherein the first angle is the angle between the speaker ray and the positive direction of the y-axis, the speaker ray is the ray formed by the actual listening position pointing to the speaker, the positive direction of the y-axis is the direction from the coordinate origin to the center speaker, and the second angle is the horizontal angle of the speaker ray; then the vocal width coefficient, the first angle, the second angle and the listening distance of each speaker are input into the first vocal constraint model for solution to obtain the vocal gain of each speaker, and the specific steps of determining the vocal gain of each speaker based on the first vocal constraint model, the vocal width coefficient of each speaker and the listening distance are clarified. Through the data such as the listening distance, the first angle and the second angle, the relative position of the actual listening position and each speaker is defined from multiple dimensions, so that the vocal gain of the speaker is calculated based on the relative position of the actual listening position and each speaker and the vocal width coefficient, which not only allows the listeners in different positions to obtain the same immersive experience, but also avoids widening the vocal signal in the music, thereby improving the listening effect.

[0103] Specifically, the first vocal constraint model includes a first constraint formula, a second constraint formula and a third constraint formula. The vocal width coefficient, the first angle, the second angle and the listening distance of each loudspeaker are input into the first vocal constraint model for solving to obtain the vocal gain of each loudspeaker, including the following steps:

[0104] S2031: adjusting the symmetrical speaker groups in the target sound system according to the actual listening position to obtain a plurality of corresponding speaker groups.

[0105] At this time, it is necessary to first obtain multiple symmetrical speaker groups in the target sound system. As mentioned above, the positive direction of the y-axis is the direction from the origin of the coordinate system to the position of the center speaker. The speakers in the symmetrical speaker group are symmetrical with respect to the yz plane. Figure 6 As shown, the multiple symmetrical speaker groups include a main symmetrical speaker group, a surround symmetrical speaker group, a rear symmetrical speaker group, and one or two symmetrical speaker groups corresponding to the sky speakers. Then, according to the actual listening position, it is determined whether the symmetrical speaker groups in the target sound system need to be adjusted: when the actual listening position is on the y-axis, no adjustment is required, and the multiple symmetrical speaker groups are directly recorded as multiple corresponding speaker groups; when the actual listening position deviates from the y-axis, the center speaker is added to the main symmetrical speaker group as a new corresponding speaker group, and then combined with other symmetrical speaker groups to obtain multiple corresponding speaker groups. Among them, the new corresponding speaker group includes (center speaker + speaker located next to the center speaker and away from the actual listening position), speaker located next to the center speaker and close to the actual listening position, that is, the center speaker + speaker located next to the center speaker and away from the actual listening position is corresponded to the speaker located next to the center speaker and close to the actual listening position. As shown Figure 6 As shown, if the actual listening position deviates from the y-axis and the actual listening position is on the left side of the y-axis ( Figure 6 ), the center speaker is added to the main symmetrical speaker group (including the left speaker L and the speaker R) as a new corresponding speaker group, and the new corresponding speaker group includes the corresponding speakers of (C+R) and L.

[0106] S2032: Input the first angle and listening distance of each speaker into the first constraint formula, input the second angle and listening distance of each speaker in the corresponding speaker group into the second constraint formula, input the vocal width coefficient and listening distance of each speaker into the third constraint formula, and solve to obtain the vocal gain of each channel.

[0107] By inputting the first angle and listening distance of each speaker into the first constraint formula, inputting the second angle and listening distance of each speaker in the corresponding speaker group into the second constraint formula, and inputting the vocal width coefficient and listening distance of each speaker into the third constraint formula, the vocal gain of each channel can be solved.

[0108] According to the placement of the loudspeaker and the actual listening position, in order to obtain a better listening effect, the first constraint formula can be obtained according to the vector superposition principle of the sound signal. The first constraint formula is expressed as follows:

[0109]

[0110]

[0111] Where n represents the nth loudspeaker; θ n represents the first angle of the loudspeaker n, that is, the angle formed by the ray formed by the actual listening position pointing to the placement position of the loudspeaker n and the positive direction of the y-axis; P n represents the vocal gain of speaker n; L n It represents the listening distance of speaker n, that is, the straight-line distance between speaker n and the actual listening position. represents the component of the audio signal of speaker n in the xy plane;

[0112] Represents the energy of the audio signal of speaker n.

[0113] In order to ensure that the sound effects on the left and right sides of the listener are consistent, based on the principle of symmetry between the left and right sides of the listener position (actual listening position), a second constraint formula constraining each corresponding speaker group can be obtained. The second constraint formula is expressed as follows:

[0114]

[0115] Among them, P left With P rig h t θ are the channel gains of the two speakers in a corresponding speaker group, which can be understood as the vocal gain of the speaker on the left side of the actual listening position and the vocal gain of the speaker on the right side of the actual listening position in a corresponding speaker group; left With θ rig h t L is the second angle of two speakers in a corresponding speaker group, that is, the horizontal angle of the speaker on the left side of the actual listening position and the horizontal angle of the speaker on the right side of the actual listening position; 1left With L rig h t They are respectively the listening distances of two speakers in a corresponding speaker group, that is, the straight-line distance between the speaker on the left side of the actual listening position and the actual listening position, and the straight-line distance between the speaker on the right side of the actual listening position and the actual listening position.

[0116] Among them, since the sound field width coefficient can affect the sound field width of the sound, thereby affecting the sound effect, the sound field width formed by the vocal signal can be flexibly controlled by the vocal width coefficient. According to the listening preferences of different listeners, the vocal width coefficient can be used to adjust the vocal gain of each speaker, thereby obtaining the third constraint formula for constraining each speaker group. The third constraint formula is expressed as follows:

[0117]

[0118] Where n represents the nth loudspeaker; θ n represents the first angle of the loudspeaker n; P n represents the vocal gain of speaker n; L n represents the listening distance of speaker n; a n Indicates the vocal width coefficient of each speaker n.

[0119] In this embodiment, the first vocal constraint model includes a first constraint formula, a second constraint formula and a third constraint formula; the symmetrical speaker group in the target sound system is adjusted according to the actual listening position to obtain multiple corresponding speaker groups, the first angle and the listening distance of each speaker are input into the first constraint formula, the second angle and the listening distance of each speaker in the corresponding speaker group are input into the second constraint formula, the vocal width coefficient and the listening distance of each speaker are input into the third constraint formula, and the vocal gain of each channel is obtained by solving the problem, which clarifies the specific process of calculating the vocal gain of each speaker and provides a data basis for the subsequent distribution of the audio signals of the speakers.

[0120] In one embodiment, in order to ensure that the sound effects on the left and right sides of the listener are similar, based on the principle of consistent sound effects on the same side and based on the listening preferences of different listeners, it is necessary to use a vocal width coefficient to constrain the speakers located on the same side of the actual listening position to obtain a better listening experience. That is, after adjusting the symmetrical speaker group in the target sound system according to the actual listening position to obtain multiple corresponding speaker groups, it is necessary to group the multiple speakers on the left and right sides according to the actual listening position, record the multiple speakers on the same side of the actual listening position as the same-side speaker group, and then input the first angle and listening distance of each speaker into the first constraint formula, input the second angle and listening distance of each speaker in the corresponding speaker group into the second constraint formula, and input the vocal width coefficient and listening distance of each speaker in the same-side speaker group into the third constraint formula to solve and obtain the vocal gain of each channel.

[0121] The human voice width coefficient is used to constrain each speaker in the same-side speaker group, thereby obtaining a third constraint formula for constraining the same-side speaker group. The third constraint formula can also be expressed as follows:

[0122]

[0123] Where j represents the jth speaker in the same-side speaker group, that is, the jth speaker located on the same side of the actual listening position, and j is less than n; θ j represents the first angle of the loudspeaker j in the loudspeaker group on the same side; P j L represents the vocal gain of speaker j in the speaker group on the same side; j represents the listening distance of speaker j in the speaker group on the same side; a j Represents the vocal width coefficient of speaker j in the same-side speaker group.

[0124] Wherein, the vocal width coefficient is a preset constant vector. Wherein, the vocal width coefficient and the accompaniment width coefficient are coordinated with each other, so as to avoid widening the vocal signal in the music and only widen the accompaniment signal. The vocal width coefficients corresponding to the same-side speaker groups on different sides of the actual listening position can be the same or different, which is determined according to the user's hearing preference. Wherein, the vocal width coefficients of each speaker in the same same-side speaker group can be the same or different, which is determined according to the user's hearing preference. When the vocal width coefficients of each speaker in the same same-side speaker group are different, the vocal width coefficients of each speaker in the same-side speaker group decrease as the distance between the speaker and the center speaker increases, which will make the subjective listening experience of the subsequently generated multi-channel audio file better.

[0125] Among them, only the multiple speakers in the first speaker group can be grouped on the left and right sides to obtain two same-side speaker groups. The second speaker group (i.e., the sky speaker group) does not apply the sound field width coefficient (human voice width coefficient and accompaniment width coefficient), so there is no need to group on the left and right sides. That is, for the first speaker group, the symmetrical speaker group in the target sound system is adjusted according to the actual listening position to obtain multiple corresponding speaker groups, and the multiple speakers are grouped on the left and right sides according to the actual listening position. The multiple speakers located on the same side of the actual listening position are recorded as the same-side speaker group, and then the first angle and listening distance of each speaker are input into the first constraint formula, the second angle and listening distance of each speaker in the corresponding speaker group are input into the second constraint formula, and the human voice width coefficient and listening distance of each speaker in the same-side speaker group are input into the third constraint formula, and the human voice gain of each channel is obtained by solving.

[0126] Among them, taking the 7.1.4 channel sound system as an example, Figure 6 and Figure 7As shown, L, Ls, Lb, Ltf, and Ltb are all on the same side, and can be a speaker group on the same side; R, Rs, Rb, Rtf, and Rtb are all on the same side, and can be a speaker group on the same side. Since Ltf, Ltb, Rtf, and Rtb are sky channels, for the first speaker group, the two speaker groups on the same side are L, Ls, and Lb; R, Rs, and Rb. At this time, the vocal width coefficients of L, Ls, and Lb are represented by a1, a2, and a3, respectively, and a1, a2, and a3 can be the same or different. According to the test results, when a1=2a2=3a3, the generated multi-channel audio file has a better subjective listening experience.

[0127] In this embodiment, a plurality of corresponding speaker groups are obtained by adjusting the symmetrical speaker group in the target sound system according to the actual listening position, and the plurality of speakers are grouped into left and right sides according to the actual listening position, and the plurality of speakers located on the same side of the actual listening position are recorded as the same-side speaker group, and then the first angle and the listening distance of each speaker are input into the first constraint formula, the second angle and the listening distance of each speaker in the corresponding speaker group are input into the second constraint formula, and the vocal width coefficient and the listening distance of each speaker in the same-side speaker group are input into the third constraint formula, and the vocal gain of each channel is obtained by solving the formula; another specific process of calculating the vocal gain of each speaker is clarified, and the vocal width coefficient is used to constrain the speakers located on the same side of the actual listening position, so as to obtain a better listening experience.

[0128] In one embodiment, in step S2223, that is, for the second speaker group, based on the second vocal constraint model and the listening distance of each speaker, determining the vocal gain of each speaker specifically includes the following steps:

[0129] S22231: Determine a first angle of each speaker.

[0130] The process of determining the first angle of the speaker is as described above and will not be repeated here.

[0131] S22232: Input the first angle and the listening distance of each speaker in the second speaker group into the second vocal constraint model for solving, so as to obtain the vocal gain of each speaker in the second speaker group.

[0132] Then, the first angle of each speaker in the second speaker group and the listening distance of each speaker are input into the second vocal constraint model for solution to obtain the vocal gain of each speaker in the second speaker group. That is, the first angle and listening distance of each sky speaker are input into the second vocal constraint model for solution to obtain the vocal gain of each sky speaker. Among them, considering that the sky speaker affects the panoramic sound effect, there is no need to adjust the sound field width, and considering the influence of the height of the sky speaker on the sound effect, the fourth constraint formula, i.e., the second vocal constraint model, is obtained by constraining each sky speaker.

[0133] When the number of sky speakers (represented by m) is 4, that is, the second speaker group includes 4 speakers (sky speakers), m=1, 2, 3, 4, the second vocal constraint model is as follows:

[0134]

[0135]

[0136] Among them, θ1, θ2, θ3, θ4 represent the first angles of the 1st to 4th sky speakers respectively; P1, P2, P3, P4 represent the vocal gains of the 1st to 4th sky speakers respectively; L1, L2, L3, L4 represent the listening distances of the 1st to 4th sky speakers respectively. Among them, the 1st sky speaker and the 2nd sky speaker form a corresponding speaker group; the 3rd sky speaker and the 4th sky speaker form a corresponding speaker group. The 1st sky speaker and the 3rd sky speaker are located on the same side, forming a same-side speaker group; the 2nd sky speaker and the 2nd sky speaker are located on the same side, forming a same-side speaker group. Reference Figure 7 The first to fourth sky speakers can be the upper left front speaker group Ltf, the upper right front speaker group Rtf, the upper left rear speaker group Ltb, and the upper right rear speaker group Rtb respectively.

[0137] When the number of the sky speakers is 2, that is, the second speaker group includes two speakers (sky speakers), the second vocal constraint model is as follows:

[0138]

[0139]

[0140] Among them, θ1 and θ2 represent the first angle of the first sky speaker and the first angle of the fourth sky speaker respectively; P1 and P2 represent the vocal gain of the first sky speaker and the channel gain of the fourth sky speaker respectively; L1 and L2 represent the listening distance of the first sky speaker and the listening distance of the fourth sky speaker respectively.

[0141] In other embodiments, the process of determining the vocal gain of each speaker in the second speaker group may also refer to the calculation process of the vocal gain of each speaker in the first speaker group in the foregoing text, which will not be repeated here.

[0142] In this embodiment, by determining the first angle of each speaker, and then directly inputting the first angle and the listening distance of each speaker in the second speaker group into the second vocal constraint model for solving, the vocal gain of each speaker in the second speaker group is obtained, and the specific process of determining the vocal gain of each speaker in the second speaker group is clarified, providing an accurate data basis for subsequent calculations.

[0143] It should be understood that the calculation process of the accompaniment gain of each speaker is similar to the calculation process of the vocal gain. Specifically, in step S223, the accompaniment gain of each speaker is determined according to the accompaniment width coefficient and the listening distance of each speaker, which specifically includes the following steps:

[0144] S2221: When there is no overhead speaker in the target sound system, determine the accompaniment gain of each speaker based on the first accompaniment constraint model, the vocal width coefficient of the speaker, and the listening distance.

[0145] After determining the sound field width coefficient of the speaker, determine whether there is a sky speaker in the target sound system. When it is determined that there is no sky speaker in the target sound system, there is no need to consider the differences in sky speakers. At this time, the accompaniment gain of each speaker can be determined directly based on the first vocal constraint model, the accompaniment width coefficient of each speaker, and the straight-line distance between each speaker and the actual listening position (ie, the listening distance).

[0146] S2232: When the target sound system has a sky speaker, divide the plurality of speakers into a first speaker group and a second speaker group.

[0147] S2233: For the first speaker group, determine the accompaniment gain of each speaker based on the first accompaniment constraint model, the vocal width coefficient of each speaker, and the listening distance.

[0148] S2234: For the second speaker group, determine the accompaniment gain of each speaker based on the second accompaniment constraint model and the listening distance of each speaker.

[0149] Specifically, for the second speaker group, based on the second accompaniment constraint model and the listening distance of each speaker, determining the accompaniment gain of each speaker includes the following steps:

[0150] S22231: Determine a first angle of each speaker.

[0151] The process of determining the first angle of the speaker is as described above and will not be repeated here.

[0152] S22232: Input the first included angle and the listening distance of each speaker in the second speaker group into the second accompaniment constraint model for solving, so as to obtain the accompaniment gain of each speaker in the second speaker group.

[0153] Then, the first angle of each speaker in the second speaker group and the listening distance of each speaker are input into the second accompaniment constraint model for solution, and the accompaniment gain of each speaker in the second speaker group is obtained. That is, the first angle and listening distance of each sky speaker are input into the second vocal constraint model for solution, and the vocal gain of each sky speaker is obtained. Among them, considering that the sky speaker affects the panoramic sound effect, there is no need to adjust the width of the sound field, and considering the influence of the height of the sky speaker on the sound effect, the fourth constraint formula, i.e., the second accompaniment constraint model, is obtained by constraining each sky speaker. In this embodiment, the second accompaniment constraint model is the same as the second identical constraint model.

[0154] When the number of the sky speakers is 4, that is, the second speaker group includes 4 speakers (sky speakers), the second accompaniment constraint model is as follows:

[0155]

[0156]

[0157] Among them, θ1, θ2, θ3, θ4 represent the first angles of the 1st to 4th sky speakers respectively; P1, P2, P3, P4 represent the accompaniment gains of the 1st to 4th sky speakers respectively; L1, L2, L3, L4 represent the listening distances of the 1st to 4th sky speakers respectively. Among them, the 1st sky speaker and the 2nd sky speaker form a corresponding speaker group; the 3rd sky speaker and the 4th sky speaker form a corresponding speaker group. The 1st sky speaker and the 3rd sky speaker are located on the same side, forming a same-side speaker group; the 2nd sky speaker and the 2nd sky speaker are located on the same side, forming a same-side speaker group. Reference Figure 7 The first to fourth sky speakers can be the upper left front speaker group Ltf, the upper right front speaker group Rtf, the upper left rear speaker group Ltb, and the upper right rear speaker group Rtb respectively.

[0158] When the number of the sky speakers is 2, that is, the second speaker group includes two speakers (sky speakers), the second accompaniment constraint model is as follows:

[0159]

[0160]

[0161] Among them, θ1 and θ2 represent the first angle of the first sky speaker and the first angle of the fourth sky speaker respectively; P1 and P2 represent the accompaniment gain of the first sky speaker and the channel gain of the fourth sky speaker respectively; L1 and L2 represent the listening distance of the first sky speaker and the listening distance of the fourth sky speaker respectively.

[0162] In other embodiments, the process of determining the accompaniment gain of each speaker in the second speaker group may also refer to the calculation process of the accompaniment gain of each speaker in the first speaker group in the foregoing text, which will not be repeated here.

[0163] In this embodiment, by determining the first angle of each speaker, and then directly inputting the first angle and the listening distance of each speaker in the second speaker group into the second accompaniment constraint model for solving, the accompaniment gain of each speaker in the second speaker group is obtained, and the specific process of determining the accompaniment gain of each speaker in the second speaker group is clarified, providing an accurate data basis for subsequent calculations.

[0164] In one embodiment, in step S2221 or S2223, the vocal gain of each speaker is determined based on the first accompaniment constraint model, the vocal width coefficient of each speaker, and the listening distance, and specifically includes the following steps:

[0165] S211: Determine a first angle and a second angle of each speaker.

[0166] When calculating the channel gain of each speaker, it is first necessary to determine the first angle and the second angle of each speaker according to the actual listening position and the placement position of each speaker.

[0167] S212: Inputting the vocal width coefficient, the first angle, the second angle and the listening distance of each loudspeaker into the first accompaniment constraint model for solving, thereby obtaining the accompaniment gain of each loudspeaker.

[0168] In this embodiment, by determining the first angle and the second angle of each speaker, the vocal width coefficient, the first angle, the second angle and the listening distance of each speaker are input into the first accompaniment constraint model for solution, and the accompaniment gain of each speaker is obtained, and the specific steps of determining the vocal gain of each speaker based on the first accompaniment constraint model, the vocal width coefficient of each speaker and the listening distance are clarified. Through data such as the listening distance, the first angle and the second angle, the relative position of the actual listening position and each speaker is defined from multiple dimensions, so that the accompaniment gain of the speaker is calculated based on the relative position of the actual listening position and each speaker and the vocal width coefficient, which not only allows people in different positions to obtain the same immersive experience, but also avoids widening the vocal signal in the music, thereby improving the listening effect.

[0169] Specifically, the first accompaniment constraint model includes a fifth constraint formula, a second constraint formula and a sixth constraint formula. The vocal width coefficient, the first angle, the second angle and the listening distance of each speaker are input into the first accompaniment constraint model for solving to obtain the accompaniment gain of each speaker. The accompaniment gain calculation process of each speaker includes the following steps:

[0170] S2121: Acquire multiple corresponding speaker groups in the target sound system.

[0171] The process of determining the multiple corresponding speaker groups is as described above and will not be repeated here.

[0172] S2122: Input the first angle and listening distance of each speaker into the first constraint formula, input the second angle and listening distance of each speaker in the corresponding speaker group into the second constraint formula, input the vocal width coefficient and listening distance of each speaker into the third constraint formula, and solve to obtain the accompaniment gain of each channel.

[0173] As mentioned above, the second constraint formula is expressed as follows:

[0174]

[0175] Among them, P left With P rig h t are the accompaniment gains of the two speakers in a corresponding speaker group; θ left With θ rig h t are the second included angles of two speakers in a corresponding speaker group, that is; L 1left With L rig h t They are respectively the listening distances of the two speakers in a corresponding speaker group.

[0176] In order to obtain a better listening effect, according to the vector superposition principle of sound signals, the fifth constraint formula can be obtained, and the first constraint formula is expressed as follows:

[0177]

[0178]

[0179] Where n represents the nth loudspeaker; θ n represents the first angle of the loudspeaker n; P n represents the accompaniment gain of speaker n; L n Indicates the listening distance of speaker n. Represents the components of the audio signal of speaker n in the xy plane.

[0180] Among them, since the sound field width coefficient can affect the sound field width of the sound, thereby affecting the sound effect, according to the listening preferences of different listeners, the accompaniment width coefficient can be used to adjust the accompaniment gain of each speaker, thereby obtaining the sixth constraint formula for constraining each speaker group. The sixth constraint formula is expressed as follows:

[0181]

[0182] Where n represents the nth loudspeaker; θ n represents the first angle of the loudspeaker n; P n represents the accompaniment gain of speaker n; L n represents the listening distance of speaker n; b n Indicates the accompaniment width coefficient of each speaker n.

[0183] In one embodiment, in order to ensure that the sound effects on the left and right sides of the listener are similar and obtain a better listening experience, after obtaining multiple corresponding speaker groups, it is also necessary to determine the same-side speaker group in the target sound system (the determination process is described above), and then input the first angle and listening distance of each speaker into the fifth constraint formula, input the second angle and listening distance of each speaker in the corresponding speaker group into the second constraint formula, and input the accompaniment width coefficient and listening distance of each speaker in the same-side speaker group into the sixth constraint formula to solve and obtain the accompaniment gain of each channel.

[0184] At this time, only the accompaniment width coefficient is used to constrain each speaker in the same-side speaker group, thereby obtaining the sixth constraint formula for constraining the same-side speaker group. The third constraint formula is expressed as follows:

[0185]

[0186] Where, j represents the jth loudspeaker; θ j represents the first angle of loudspeaker j; P j represents the accompaniment gain of speaker j; L j represents the listening distance of speaker j; b j Indicates the accompaniment width coefficient of each speaker j.

[0187] Wherein, the accompaniment width coefficient is a preset constant vector. Wherein, the vocal width coefficient and the accompaniment width coefficient are coordinated with each other to avoid widening the vocal signal in the music and only widen the accompaniment signal. The accompaniment width coefficients corresponding to the same-side speaker groups on different sides of the actual listening position can be the same or different, depending on the user's listening preferences. Wherein, the accompaniment width coefficients of each speaker in the same same-side speaker group can be the same or different, depending on the user's listening preferences.

[0188] In one embodiment, step S30, that is, generating an audio signal corresponding to a speaker based on the vocal signal, the accompaniment sound signal and the channel gain of each speaker, specifically includes the following steps:

[0189] S31: Determine a channel gain of a woofer and a subwoofer in a target sound system.

[0190] After determining the channel gain of each speaker, it is necessary to determine the woofer in the target sound system and determine the channel gain of the woofer (subwoofer).

[0191] S32: low-pass filtering the accompaniment sound signal to obtain a filtered accompaniment sound signal, and distributing the channel gain of the woofer and the filtered accompaniment sound signal to the woofer to obtain an audio signal of the woofer.

[0192] Then, the accompaniment sound signal is input into a low-pass filter for low-pass filtering to obtain the filtered accompaniment sound signal, and the channel gain of the woofer and the filtered accompaniment sound signal are distributed to the woofer to obtain the audio signal of the woofer. For the bass channel, the purpose is to restore the low-frequency soundtrack (accompaniment) such as drums and bass in the music as much as possible, and step S10 has separated the human voice and the accompaniment for the stereo audio, so it is only necessary to perform low-pass filtering on the separated accompaniment sound, and then assign the woofer with the channel gain of the woofer, without assigning the human voice signal. Wherein, in this embodiment, the channel gain of the woofer can only include the accompaniment gain, and the accompaniment gain of the woofer and the filtered accompaniment sound signal are directly assigned to the woofer, which can reduce the calculation amount of the human voice gain of the woofer. In other embodiments, the channel gain of the woofer can be a fixed value (for example, it can be 0), and then the fixed value is used as the channel gain of the woofer, and is distributed to the woofer together with the filtered accompaniment sound signal, further reducing the calculation amount required for the sound gain, reducing the load and generating speed through multi-channel audio files.

[0193] S33: Distribute the vocal signal, the accompaniment sound signal and the channel gains of the remaining speakers to the corresponding speakers to obtain the audio signals of the corresponding speakers.

[0194] Finally, the vocal signal, accompaniment signal and the channel gains of the remaining speakers are distributed to the corresponding speakers to obtain the audio signals of the corresponding speakers, and then the audio signals of all speakers are combined to obtain a multi-channel audio file.

[0195] In this embodiment, by determining the woofer and its channel gain in the target sound system, the accompaniment sound signal is low-pass filtered to obtain the filtered accompaniment sound signal, and the channel gain of the woofer and the filtered accompaniment sound signal are allocated to the woofer to obtain the audio signal of the woofer, and at the same time, the human voice signal, the accompaniment sound signal and the channel gain of the remaining speakers are allocated to the corresponding speakers to obtain the audio signal of the corresponding speakers, and the specific steps of generating the audio signal of the corresponding speaker based on the human voice signal, the accompaniment sound signal and the channel gain of the speaker are clarified. Directly allocating the accompaniment gain of the woofer and the filtered accompaniment sound signal to the woofer can reduce the calculation amount of the channel gain of the woofer while ensuring the good sound effect of the bass channel. In other embodiments, low-pass filtering may not be performed.

[0196] It should be understood that the order of execution of the steps in the above embodiment does not necessarily mean the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiment of the present invention.

[0197] In one embodiment, a multi-channel audio file generation device is provided, and the multi-channel audio file generation device corresponds to the multi-channel audio file generation method in the above embodiment. Figure 8 As shown, the multi-channel audio file generation device includes a separation module 801, a determination module 802 and a generation module 803. The detailed description of each functional module is as follows:

[0198] The separation module 801 is used to separate the human voice from the input stereo audio file to obtain a human voice signal and an accompaniment sound signal;

[0199] A determination module 802 is used to determine the arrangement positions of multiple speakers in the target sound system, and determine the channel gain of each speaker according to the arrangement positions of the multiple speakers;

[0200] The generating module 803 is used to generate audio signals corresponding to the speakers based on the vocal signal, the accompaniment sound signal and the channel gain of the speakers, and merge them into a multi-channel audio file adapted to the target sound system.

[0201] Optionally, the determining module 802 is specifically configured to:

[0202] An actual listening position is set in the target sound system, and the listening distance of each speaker is determined according to the actual listening position and the arrangement position of the speaker, where the listening distance is the straight-line distance between the speaker and the actual listening position;

[0203] The channel gain of each speaker is determined according to the listening distance of each speaker.

[0204] Optionally, the channel gain includes a vocal gain and an accompaniment gain, and the determination module 802 is further specifically used for:

[0205] Determine the sound field width coefficient of each loudspeaker, the sound field width coefficient includes a human voice width coefficient and an accompaniment width coefficient;

[0206] Determine the vocal gain of each speaker according to the vocal width coefficient of each speaker and the listening distance;

[0207] The accompaniment gain of each speaker is determined according to the accompaniment width coefficient of each speaker and the listening distance.

[0208] Optionally, the determining module 802 is further configured to:

[0209] When the target sound system has a sky speaker, the plurality of speakers are divided into a first speaker group and a second speaker group, wherein the second speaker group consists of the sky speakers;

[0210] For the first speaker group, determining a vocal gain of each speaker based on the first vocal constraint model, a vocal width coefficient of each speaker, and a listening distance;

[0211] For the second speaker group, the vocal gain of each speaker is determined based on the second vocal constraint model and the listening distance of each speaker, and the constraint formula of the second vocal constraint model is different from that of the first vocal constraint model.

[0212] Optionally, the determining module 802 is further configured to:

[0213] Determine a first angle of each speaker, where the first angle is the angle between the speaker ray and the positive direction of the y-axis, where the speaker ray is a ray formed from the actual listening position to the speaker, and the positive direction of the y-axis is the direction from the coordinate origin to the center speaker;

[0214] Determine a second angle of each loudspeaker, where the second angle is a horizontal angle of the loudspeaker ray;

[0215] The vocal width coefficient, the first angle, the second angle and the listening distance of each loudspeaker are input into the first vocal constraint model for solving to obtain the vocal gain of each loudspeaker.

[0216] Optionally, the generating module 803 is specifically used for:

[0217] Determine the channel gains of the woofers and subwoofers in the target sound system;

[0218] Performing low-pass filtering on the accompaniment sound signal to obtain a filtered accompaniment sound signal, and distributing the channel gain of the woofer and the filtered accompaniment sound signal to the woofer to obtain an audio signal of the woofer;

[0219] The vocal signal, the accompaniment sound signal and the channel gains of the remaining speakers are distributed to the corresponding speakers to obtain the audio signals of the corresponding speakers.

[0220] Optionally, the separation module 801 is specifically used for:

[0221] Obtaining a pre-trained stereo separation model, the stereo separation model comprising a signal transformation layer and a signal filter;

[0222] The stereo audio file is input into the stereo separation model, and the Fourier transform is performed through the signal transformation layer to obtain the transformed audio signal, and the human voice separation is performed on the transformed audio signal through the signal filter to obtain the human voice signal and the accompaniment sound signal.

[0223] The specific definition of the multi-channel audio file generation device can be found in the definition of the multi-channel audio file generation method above, which will not be repeated here. Each module in the above-mentioned multi-channel audio file generation device can be implemented in whole or in part by software, hardware and a combination thereof. The above-mentioned modules can be embedded in or independent of the processor in the computer device in the form of hardware, or can be stored in the memory of the computer device in the form of software, so that the processor can call and execute the operations corresponding to the above modules.

[0224] In one embodiment, a computer device is provided, which may be a server. The computer device includes a processor, a memory, a network interface and a database connected via a system bus. The processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, a computer program and a database. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The database of the computer device is used to store data used and generated by the multi-channel audio file generation method, such as a stereo separation model, a stereo audio file, a separated human voice signal and an accompaniment sound signal, a multi-channel audio file, etc. The network interface of the computer device is used to communicate with an external terminal via a network connection. When the computer program is executed by the processor, a multi-channel audio file generation method is implemented.

[0225] In one embodiment, Figure 8 As shown, 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 the above-mentioned multi-channel audio file generation method are implemented.

[0226] In one embodiment, a computer-readable storage medium is provided, on which a computer program is stored. When the computer program is executed by a processor, the steps of the above-mentioned multi-channel audio file generation method are implemented.

[0227] A person skilled in the art can understand that all or part of the processes in the above-mentioned embodiments can be implemented 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 in this application can include non-volatile and / or volatile memory.

[0228] 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.

[0229] 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 multi-channel audio file, characterized in that: include: Separate the input stereo audio file into human voice to obtain human voice signal and accompaniment sound signal; Determining the arrangement positions of a plurality of speakers in the target sound system, and determining the channel gain of each speaker according to the arrangement positions of the plurality of speakers; An audio signal corresponding to the loudspeaker is generated based on the vocal signal, the accompaniment sound signal and the channel gain of the loudspeaker, and the audio signals are combined into a multi-channel audio file adapted to the target sound system.

2. The method for generating a multi-channel audio file according to claim 1, wherein: The step of determining the channel gain of each of the speakers according to the arrangement positions of the plurality of speakers comprises: An actual listening position is set in the target sound system, and a listening distance of each of the speakers is determined according to the actual listening position and the arrangement position of the speakers, wherein the listening distance is a straight-line distance between the speakers and the actual listening position; The channel gain of each of the speakers is determined according to the listening distance of each of the speakers.

3. The method for generating a multi-channel audio file according to claim 2, wherein: The channel gain includes a vocal gain and an accompaniment gain, and the determining of the channel gain of each speaker according to the listening distance of each speaker includes: Determining a sound field width coefficient of each of the loudspeakers, wherein the sound field width coefficient includes a vocal width coefficient and an accompaniment width coefficient; Determining the vocal gain of each of the speakers according to the vocal width coefficient of each of the speakers and the listening distance; The accompaniment gain of each of the speakers is determined according to the accompaniment width coefficient of each of the speakers and the listening distance.

4. The method for generating a multi-channel audio file according to claim 3, wherein: The step of determining the vocal gain of each speaker according to the vocal width coefficient of each speaker and the listening distance includes: When the target sound system has a sky speaker, the plurality of speakers are divided into a first speaker group and a second speaker group, wherein the second speaker group is composed of the sky speakers; For the first speaker group, determining a vocal gain of each speaker based on a first vocal constraint model, the vocal width coefficient of each speaker, and the listening distance; For the second speaker group, the vocal gain of each speaker is determined based on a second vocal constraint model and a listening distance of each speaker, and the constraint formula of the second vocal constraint model is different from that of the first vocal constraint model.

5. The method for generating a multi-channel audio file according to claim 4, wherein: The determining the vocal gain of each of the loudspeakers based on the first vocal constraint model, the vocal width coefficient of each of the loudspeakers, and the listening distance includes: Determine a first angle of each of the speakers, where the first angle is an angle between a speaker ray and a positive direction of the y-axis, the speaker ray is a ray formed from the actual listening position to the speaker, and the positive direction of the y-axis is a direction from the coordinate origin to the center speaker; Determine a second angle of each of the loudspeakers, where the second angle is a horizontal angle of a loudspeaker ray; The vocal width coefficient, the first angle, the second angle and the listening distance of each of the loudspeakers are input into the first vocal constraint model for solution to obtain the vocal gain of each of the loudspeakers.

6. The method for generating a multi-channel audio file according to any one of claims 1 to 5, characterized in that: The step of generating an audio signal corresponding to the loudspeaker based on the vocal signal, the accompaniment sound signal and the channel gain of the loudspeaker comprises: determining a woofer in the target sound system and a channel gain of the woofer; Low-pass filtering the accompaniment sound signal to obtain the filtered accompaniment sound signal, and distributing the channel gain of the woofer and the filtered accompaniment sound signal to the woofer to obtain an audio signal of the woofer; The vocal signal, the accompaniment sound signal and the channel gains of the remaining speakers are distributed to the corresponding speakers to obtain audio signals corresponding to the speakers.

7. The method for generating a multi-channel audio file according to any one of claims 1 to 5, characterized in that: The step of separating the input stereo audio file into human voices to obtain a human voice signal and an accompaniment sound signal comprises: Acquire a pre-trained stereo separation model, wherein the stereo separation model includes a signal transformation layer and a signal filter; The stereo audio file is input into the stereo separation model, Fourier transform is performed through the signal transformation layer to obtain a transformed audio signal, and human voice separation is performed on the transformed audio signal through the signal filter to obtain the human voice signal and the accompaniment sound signal.

8. A multi-channel audio file generating device, characterized in that: include: The separation module is used to separate the human voice from the input stereo audio file to obtain a human voice signal and an accompaniment sound signal; A determination module, used to determine the arrangement positions of a plurality of loudspeakers in a target sound system, and determine the channel gain of each of the loudspeakers according to the arrangement positions of the plurality of loudspeakers; A generating module is used to generate an audio signal corresponding to the loudspeaker based on the vocal signal, the accompaniment sound signal and the channel gain of the loudspeaker, and merge the audio signals into a multi-channel audio file adapted to the target sound system.

9. 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 a multi-channel audio file according to any one of claims 1 to 7 are implemented.

10. 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 a multi-channel audio file according to any one of claims 1 to 7 are implemented.

Images