Disclosure of Invention
The invention mainly aims to provide a multifunctional integrated song ordering machine and aims to solve the defect that an intelligent accompaniment function does not support a multi-user chorus scene.
In order to achieve the above object, the multifunctional integrated song ordering machine provided by the present invention is used for being connected with N microphones and playing devices, and the multifunctional integrated song ordering machine includes:
The acquisition module is in wireless communication connection with the N microphones and is used for acquiring N paths of singing sound information respectively output by the N microphones, wherein N is more than or equal to 2;
a dividing module for obtaining accompaniment audio and original singing audio of the current on-demand song and dividing the on-demand song into a plurality of singing intervals according to the accompaniment audio and the original singing audio, and
The control module is connected with the acquisition module, the dividing module and the playing equipment, and is used for dividing the middle N paths of singing sound information into first singing information and at least one second singing information in the singing interval, and adjusting accompaniment audio and original singing audio according to the first singing information and the second singing information and outputting the accompaniment audio and the original singing audio to the playing equipment;
wherein the singing effect of the first singing information is better than that of the second singing information.
Optionally, each singing interval comprises an acquisition interval and an adjustment interval which are sequentially carried out;
for each singing interval, the control module is used for carrying out score evaluation on each path of singing voice information based on a plurality of evaluation dimensions, and determining the first singing information and the second singing information according to score evaluation results of each path of singing voice information;
The control module is further used for adjusting the accompaniment audio and the original singing audio according to the first singing information and the second singing information in the adjustment interval and outputting the accompaniment audio and the original singing audio to the playing equipment;
The ratio of the duration of the acquisition interval to the duration of the singing interval is not more than 20%.
Optionally, the control module is further configured to adjust, in the mth acquisition interval, accompaniment audio and original audio according to the first singing information and each second singing information in the previous singing interval, and output the accompaniment audio and the original singing information to the playing device;
Wherein M is more than or equal to 2.
Optionally, the plurality of evaluation dimensions are singing tempo, singing emotion and singing intonation;
the control module is used for carrying out weighted calculation on the evaluation score of the singing beat, the evaluation score of the singing emotion and the evaluation score of the singing voice level to obtain a score evaluation result of the singing voice information.
Optionally, the control module is configured to determine, as first singing information in the singing interval, one of the N paths of singing sound information with a highest evaluation score result, and determine the remaining singing sound information as second singing information.
Optionally, the control module is further configured to, when score evaluation results of the multiple paths of singing voice information are the highest and the same, determine the singing voice information with the highest sum of evaluation scores of preset evaluation dimensions as first singing information, and determine the remaining singing voice information as second singing information;
The preset evaluation dimension is the singing beat and the singing level.
Optionally, the control module is used for adjusting the output proportion of the original singing audio according to the singing beat fraction and the singing tone quasi fraction of the first singing information;
The control module is used for adjusting the output proportion of the accompaniment audio according to the singing emotion scores of the first singing information and the second singing information.
Optionally, the control module is configured to close or reduce the output magnification of the original singing audio when the sum of the singing beat score and the singing pitch score of the first singing information is in a first preset score interval;
When the singing beat score and the singing tone quasi score sum value of the first singing information are in a second preset score interval, adjusting the output multiplying power of the original singing audio to the original singing output multiplying power or increasing the output multiplying power to a first target multiplying power;
Increasing the output multiplying power of the original singing audio to a second target multiplying power under the condition that the singing beat fraction and the singing tone quasi fraction sum value of the first singing information are in a third preset fraction interval;
the first target multiplying power is larger than the initial original singing output multiplying power and smaller than the second target multiplying power, and the scores corresponding to the first preset score interval, the second preset score interval and the third preset score interval are sequentially reduced.
Optionally, the control module is configured to perform average value calculation on the singing emotion scores of the first singing information and the second singing information;
increasing the output multiplying power of the accompaniment audio to a third target multiplying power under the condition that the average value calculation result is in a fourth preset fractional interval;
increasing the output multiplying power of the accompaniment audio to a fourth target multiplying power under the condition that the average value calculation result is in a fifth preset fractional interval;
When the average value calculation result is in a sixth preset fractional interval, adjusting the output multiplying power of the accompaniment audio to an initial accompaniment output multiplying power;
The fourth target multiplying power is larger than the initial accompaniment output multiplying power and smaller than the third target multiplying power, and scores corresponding to the fourth preset score interval, the fifth preset score interval and the sixth preset score interval are sequentially reduced.
Optionally, the multifunctional integrated song ordering machine further comprises:
For the singing sound information output by any one microphone, the filtering module is used for carrying out filtering processing by utilizing the accompaniment audio, the original singing audio and the singing sound information output by other N-1 microphones and outputting the processed singing sound information to the control module;
The filtering process is used for filtering out audio components corresponding to the accompaniment audio, the original singing audio and the singing sound information output by other N-1 microphones.
According to the technical scheme, the plurality of singing intervals are divided, the first singing information and the second singing information which are most suitable at present are determined in real time in each singing interval, accompaniment audios and original singing audios in the singing interval are adjusted according to the first singing information and the second singing information, and accordingly the playing equipment can play the accompaniment audios and the original singing audios which are more suitable for the whole chorus effect of the current playing interval, the intelligent accompaniment function can be achieved under the scene with multi-person chorus, and the defect that the intelligent accompaniment function does not support the multi-person chorus scene is overcome.
Detailed Description
The following description of the embodiments of the present invention will be made clearly and fully with reference to the accompanying drawings, in which it is evident that the embodiments described are only some, but not all embodiments of the invention. All other embodiments, which can be made by those skilled in the art based on the embodiments of the invention without making any inventive effort, are intended to be within the scope of the invention.
Furthermore, descriptions such as those referred to as "first," "second," and the like, are provided for descriptive purposes only and are not to be construed as indicating or implying a relative importance or implying an order of magnitude of the indicated technical features in the present disclosure. Thus, a feature defining "a first" or "a second" may explicitly or implicitly include at least one such feature. In addition, the technical solutions of the embodiments may be combined with each other, but it is necessary to base that the technical solutions can be realized by those skilled in the art, and when the technical solutions are contradictory or cannot be realized, the combination of the technical solutions should be considered to be absent and not within the scope of protection claimed in the present invention.
Under the multi-user chorus scene, as the plurality of microphones have signal inputs at the same time, and under the multi-user chorus condition, singers generally have extremely different singing levels, according to the adjusting logic of the current intelligent accompaniment function, the intelligent accompaniment function is easy to generate the contradiction of the adjusting results of the intelligent accompaniment function according to different microphone outputs, for example, the original singing audio is adjusted according to the user microphone signal input with poor singing effect, and the adjusting result is obtained according to the user microphone signal input with good singing effect, so that the original singing audio is reduced. In order to avoid the situation that the above adjustment results are contradictory, the intelligent accompaniment function in the prior art only allows to be started in the solo scene of the user.
In order to solve the problems, the invention provides a multifunctional integrated song ordering machine, and aims to realize an intelligent accompaniment function which can be suitable for being used in a multi-person chorus scene.
Referring to fig. 1, the multifunctional integrated song ordering machine includes:
The acquisition module is in wireless communication connection with the N microphones and is used for acquiring N paths of singing sound information respectively output by the N microphones, wherein N is more than or equal to 2;
a dividing module for obtaining the accompaniment audio and the original singing audio of the current on-demand song and dividing the on-demand song into a plurality of singing intervals according to the accompaniment audio and the original singing audio, and
The control module is connected with the acquisition module, the dividing module and the playing equipment, and is used for dividing the middle N-channel singing voice information into first singing information and at least one second singing information in a singing interval, and adjusting accompaniment audio and original singing audio according to the first singing information and the second singing information and outputting the accompaniment audio and the original singing audio to the playing equipment;
Wherein, the singing effect of the first singing information is better than that of the second singing information.
In this embodiment, the multifunctional integrated song ordering machine is used to connect with N microphones and playing devices. N is determined by the number of users currently choreong, in this case N is at least 2. The acquisition module can be connected with the N microphones in a wireless communication mode such as Bluetooth communication and wifi communication, so that when different users are accessed to sing, the microphones acquire singing sound information output after singing sounds of the users. The accompaniment audio continues the whole singing process, while the original singing audio only appears when singing is needed, and the whole singing process is not maintained.
The dividing module is configured to distinguish each complete lyric according to the lyric information, then combine the original singing audio to determine the singing time period of each complete lyric, and finally use each determined singing time period as a singing interval. It will be appreciated that there is a portion of the singing audio in each singing interval.
The control module can determine one path of singing sound information with the best singing effect as first singing information according to N paths of singing sound information after entering one singing interval, and take other paths of singing sound information with non-best singing effect as second singing information, and in each singing interval, according to the first singing information and the second singing information which are determined currently, the control module adjusts output accompaniment audio and original singing audio, and outputs the adjusted accompaniment audio and original singing audio to the playing equipment for playing, so that the playing equipment can play the accompaniment audio and the original singing audio which are more in agreement with the whole chorus effect of the current playing interval. It is understood that the sum of the amounts of the first singing information and the second singing information is N.
Thus, since each chorus has parts which are good at playing singing and are more willing to sing, and the parts of the chorus are not always the same, the technical scheme of the invention determines the most suitable first singing information and the second singing information in real time in each singing interval by dividing a plurality of singing intervals, and the accompaniment audio and the original singing audio in the singing interval are adjusted according to the first singing information and the second singing information, so that the playing equipment can play the accompaniment audio and the original singing audio which are more suitable for the whole chorus effect of the current playing interval, thereby realizing the intelligent accompaniment function in a chorus scene with multiple persons, and further solving the defect that the intelligent accompaniment function does not support the chorus scene with multiple persons.
Referring to fig. 1, each singing section includes an acquisition section and an adjustment section that are sequentially performed;
for each singing interval, the control module is used for carrying out score evaluation on each path of singing voice information based on a plurality of evaluation dimensions, and determining first singing information and second singing information according to score evaluation results of each path of singing voice information;
the control module is also used for adjusting the accompaniment audio and the original singing audio according to the first singing information and the second singing information in the adjustment interval and outputting the accompaniment audio and the original singing audio to the playing equipment.
In this implementation, the singing section includes an acquisition section and an adjustment section that are sequentially performed, and the duration of the acquisition section is smaller than the duration of the adjustment section. The evaluation dimension is a dimension related to the singing effect. The control module may perform score evaluation on the singing voice information of each path according to a plurality of evaluation dimensions in the collection interval of each singing interval, and then determine the singing voice information with the best singing effect as the first singing information and use other singing voice information as the second singing information according to the final score evaluation results of the plurality of evaluation dimensions.
In an alternative embodiment, the ratio of the duration of the acquisition interval to the duration of the singing interval does not exceed 20%.
By the arrangement, the duration of the acquisition interval can be ensured not to be too long, and meanwhile, the adjustment interval can occupy a sufficient proportion in the whole singing interval, so that chorus experience of a user is ensured.
In the test, the situation that continuous occurrence is easy to occur among a plurality of singing intervals is found, and the accompaniment audio and the original singing audio are in a playing state in the collecting interval, so that if the initial output multiplying power of the accompaniment audio and the original singing audio is adopted for output in the collecting interval, the continuity of the accompaniment audio and the original singing audio in the continuous two singing intervals is easy to be poor, and the integral chorus effect of a user is very influenced.
Aiming at the problem, the control module is also used for adjusting the accompaniment audio and the original singing audio in the M-th acquisition interval according to the first singing information and the second singing information in the former singing interval and outputting the accompaniment audio and the original singing audio to the playing equipment;
Wherein M is more than or equal to 2.
In practical tests, it was found that in the case where the singing intervals continuously appear, although the singing effects of the respective users fluctuate among different singing intervals, the ranking of the overall singing effect is not greatly changed, i.e., the first singing information, the second singing information, and the large probability are the same singing sound information. Based on the above, the technical scheme of the invention inherits the first singing information and the second singing information in the last singing interval in the new acquisition interval by the control module, and adjusts the accompaniment audio and the original singing audio based on the first singing information and the second singing information in the new acquisition interval, so that continuity of the accompaniment audio and the original singing audio in the two singing intervals can be ensured to the greatest extent, and the influence on the overall chorus effect of a user is reduced.
Optionally, the plurality of evaluation dimensions are singing tempo, singing emotion, and singing intonation;
in this embodiment, the control module may obtain the original audio information of the original audio in the current collection interval, and the interval score information of the currently singed song in the current collection interval. The evaluation score of the singing beat dimension (i.e., singing beat score) can be determined by extracting the offset of the singing voice information relative to the interval original audio information, the evaluation score of the singing pitch dimension (i.e., singing pitch score) can be determined by comparing the frequency values of the singing voice information and the interval original audio information, and the evaluation score of the singing emotion dimension (i.e., singing emotion score) can be determined by comparing the musical intensities of the singing voice information and the interval music score.
It should be noted that, in the technical solution of the present invention, the purpose of score evaluation on each piece of singing sound information is not to comment on the singing effect of the user, but to screen out the first singing information closer to the original singing, and in a multi-user chorus scene, the evaluation score distortion obtained by many evaluation criteria is serious, for example, the singing volume, the singing voice domain, the singing voice color, the lyric accuracy and the like, which are often used for ordering on the singing effect of the user by a single person. In view of the above, the technical scheme of the invention can ensure that the influence of chorus scenes is small and simultaneously ensure the matching degree of the final score evaluation result and the actual singing effect and level by selecting three evaluation dimensions of singing beat, singing emotion and singing tone.
The control module is used for carrying out weighted calculation on scores of three evaluation dimensions of singing beat, singing emotion and singing tone, and obtaining a final score evaluation result of singing sound information. The weighting calculation specifically comprises the steps of carrying out sum calculation on the product of the singing beat score and a first preset weighting value, the product of the singing note score and a second preset weighting value and the product of the singing emotion score and a third preset weighting value, and taking the sum calculation result as a final score evaluation result of singing voice information. The first preset weight value, the second preset weight value, and the third preset weight value may be determined through a plurality of pre-experiments, which is not limited in this embodiment.
Optionally, the control module is configured to determine one path of singing voice information with the highest score evaluation result in the N paths of singing voice information as first singing information in a singing interval, and determine the remaining singing voice information as second singing information.
By the arrangement, the singing voice information with the best singing effect can be used as the first singing information under the condition that only one path of singing voice information has the highest final score evaluation result.
Optionally, the control module is configured to determine, when score evaluation results of the multiple paths of singing voice information are the highest and the same, the singing voice information with the highest sum of evaluation scores of preset evaluation dimensions as the first singing information, and the remaining singing voice information as the second singing information, where the preset evaluation dimensions are a singing tempo and a singing pitch.
In this embodiment, if the control module calculates that the final score evaluation results of at least two pieces of singing voice information are the highest, the control module returns to calculate who is the highest sum of the scores of the singing beats and the singing levels of the pieces of singing voice information, and uses the singing voice information with the highest sum of the scores of the singing beats and the singing levels as the first singing information. In other words, the score of the singing voice level is removed under the condition that the final score evaluation results of at least two paths of singing voice information are the highest, so that the first singing information is determined through two evaluation dimensions of the singing tempo and the singing voice level which can reflect the actual singing level of the singer.
It should be noted that, in the practical test, it is found that there is a large difference in the roles of the original audio and the accompaniment audio in the chorus scene and the solo scene. The biggest difference is that in the solo scene, as only one person sings, if the level of the singer is poor, the original singing audio and the accompaniment audio are often needed to be stronger to assist so as to present better singing effect. The chorus scenes are different, and because the chorus environment is numerous, the chorus environment is easy to have first singing information with the best singing effect, no matter how big a gap exists between the first singing information and the original singing audio, other singers with poor singing level are often more favored to singer singing along with the first singing information source as a chorus friend compared with the sound of the original singing audio. In addition, in a chorus scene, the accompaniment audio is not separately served to the singer with the best singing effect, but is required to serve the whole chorus effect.
Based on the above situation, the technical scheme of the invention configures the control module to adjust the output proportion of the original singing audio according to the singing beat fraction and the singing tone quasi fraction of the first singing information;
the control module is used for adjusting the output proportion of the accompaniment audio according to the singing emotion scores of the first singing information and the second singing information. By the arrangement, the output ratio of the original singing audio and the accompaniment audio can be matched with the overall chorus effect in real time.
Further, the control module is used for closing or reducing the output multiplying power of the original singing audio when the sum value of the singing beat score and the singing tone quasi score of the first singing information is in a first preset score interval;
Under the condition that the score of the singing beat and the score sum of the singing voice quasi-score of the first singing information are in a second preset score interval, the output multiplying power of the original singing voice frequency is adjusted to the original singing output multiplying power or is increased to a first target multiplying power;
and under the condition that the singing beat score and the singing tone quasi score sum value of the first singing information are in a third preset score interval, increasing the output multiplying power of the original singing audio to a second target multiplying power.
The first target multiplying power is larger than the initial original singing output multiplying power and smaller than the second target multiplying power, and the scores corresponding to the first preset score interval, the second preset score interval and the third preset score interval are sequentially reduced.
In this embodiment, the first preset fractional interval, the second preset fractional interval, and the third preset fractional interval may be three successively reduced fractional intervals. The control module is used for carrying out sum value calculation on the singing beat score and the singing pitch score of the first singing information, and judging in which preset score interval the sum value calculation result is.
If the judging result is that the sum value calculating result is in the first preset score interval, the singer level of the first singing information is extremely high, the singing effect is close to the original singing, at the moment, the output multiplying power of the original singing audio can be closed or reduced, the participation degree of the original singing audio is reduced to improve the participation feeling of each singer, and the singer with the first singing information with higher level does not influence the singing effect. If the judging result is that the sum value calculating result is in the second preset score interval, the singer level of the first singing information is higher, the singing effect is better, and at the moment, the output multiplying power of the original singing audio can be adjusted to the initial multiplying power or increased to the lower first target multiplying power, so that the participation degree of the original singing audio is slightly improved to improve the participation feeling and the chorus effect of each singer. If the judging result is that the sum value calculating result is in the third preset score interval, the singer level of the first singing information is general or worse, the singing effect is far different from that of the original singing, and at the moment, the output multiplying power of the original singing audio can be increased to a second higher target multiplying power so as to improve the participation degree of the original singing audio and improve the participation feeling of each singer.
In this embodiment, the first target magnification and the second target magnification may be a preset magnification value, or may be real-time magnification values calculated according to related parameters (for example, a singing beat score and a singing pitch score of the first singing information), which is not limited in this embodiment.
Further, the control module is used for calculating the average value of the singing emotion scores of the first singing information and the second singing information;
When the average value calculation result is in a fourth preset fraction interval, increasing the output multiplying power of the accompaniment audio to a third target multiplying power;
Increasing the output multiplying power of the accompaniment audio to a fourth target multiplying power under the condition that the average value calculation result is in a fifth preset fractional interval;
When the average value calculation result is in a sixth preset score interval, adjusting the output multiplying power of the accompaniment audio to the initial accompaniment output multiplying power;
The fourth target multiplying power is larger than the initial accompaniment output multiplying power and smaller than the third target multiplying power, and the scores corresponding to the fourth preset score interval, the fifth preset score interval and the sixth preset score interval are sequentially reduced.
The control module may add the singing emotion score of the first singing information and the singing emotion score of each second singing information and divide the added singing emotion score by N to obtain an average value calculation result. It should be noted that, actual tests show that the singing emotion score is proportional to the intensity of willingness of singers to sing, that is, if the average value calculation result is higher, the current willingness of each singer to actively sing is higher. Based on this, in the present embodiment, the fourth preset score interval, the fifth preset score interval, and the sixth preset score interval may also be three successively decreasing score intervals. The control module can judge which preset score interval the average value calculation result is in after the average value calculation is carried out on the singing emotion scores.
If the average value calculation result is in the fourth preset score interval, the judgment result shows that the current positive singing willingness of each singer is extremely high, higher accompaniment audios are needed to build singing atmosphere so as to improve the participation degree of the accompaniment audios to cooperatively improve the integral chorus effect, and the control module can increase the output multiplying power of the accompaniment audios to a higher third target multiplying power so as to meet the requirement of chorus on the accompaniment audios at the moment. If the average value calculation result is in the fifth preset score interval, the judgment result indicates that the current willingness degree of the singing by each singer is higher, the accompaniment audio can be slightly improved to assist in obtaining a better chorus effect, and the control module can increase the output multiplying power of the accompaniment audio to a slightly higher fourth target multiplying power so as to meet the requirement of chorus on the accompaniment audio at the moment. If the average value calculation result is in the sixth preset score interval, the judgment result indicates that the current willingness degree of active singing of each singer is general or low, the accompaniment audio is not required to be improved to assist in obtaining a better chorus effect, and the control module can adjust the output multiplying power of the accompaniment audio to the initial accompaniment output multiplying power.
In this embodiment, the third target magnification and the fourth target magnification may be a preset magnification value, or may be real-time magnification values calculated according to related parameters (for example, a singing beat score and a singing pitch score of the first singing information), which is not limited in this embodiment.
In an alternative embodiment, the control module may substitute the higher one of the singing tempo score and the singing pitch score of the first singing information, the singing tempo score and the singing pitch score of each second singing information into a corresponding preset amplification factor calculation formula to calculate to obtain a first multiple, and perform a product operation on the first multiple and the initial accompaniment output rate to obtain the third target rate. The control module may further replace a lower one of the singing tempo score and the singing pitch score of the first singing information, the singing tempo score and the singing pitch score of each second singing information with a corresponding preset amplification factor calculation formula to calculate to obtain a second multiple, and perform product operation on the second multiple and the initial accompaniment output rate to obtain a fourth target rate.
Because the commercial song ordering machine is generally smaller in space, and the echo is serious under the condition of multi-person chorus, the microphone is easy to collect the sounds of other singers and other sounds, and therefore the score evaluation result error of the voice information of each singing is larger.
With reference to fig. 2, in view of the above problems, the multifunctional integrated song ordering machine further includes:
The voice playing module is used for playing voice information output by any one microphone, and outputting the voice information to the control module after filtering processing by utilizing accompaniment audio, original singing audio and voice information output by other N-1 microphones;
the filtering process is used for filtering out audio components corresponding to the accompaniment audio, the original singing audio and the singing sound information output by other N-1 microphones.
Therefore, the singing voice information accessed by the control module can eliminate the accompanying audio, the original singing audio and the audio components corresponding to the singing voice information output by other N-1 microphones, so that the evaluation accuracy of the control module on each path of singing voice information is improved.
The foregoing description of the embodiments of the present invention is merely an optional embodiment of the present invention, and is not intended to limit the scope of the invention, and all equivalent structural modifications made by the present invention in the light of the present invention, the description of which and the accompanying drawings, or direct/indirect application in other related technical fields are included in the scope of the invention.