Karaoke equipment

Abstract

The present invention provides a karaoke device capable of determining whether or not whisper singing is included in a karaoke performance. [Solution] A karaoke device having an acquisition unit that acquires singing voice signals associated with a singer's karaoke singing at predetermined intervals, and a determination unit that determines whether the singer's karaoke singing is whisper voice singing based on the spectral contrast obtained from the singing voice signals at predetermined intervals acquired by the acquisition unit.

Description

Technical Field

[0001] The present invention relates to a karaoke device.

Background Art

[0002] Among singers who perform karaoke singing using a karaoke device, there are those who imitate the singing of professional singers and sing using special singing techniques such as sobbing, falling, punching, and shouting. Patent Documents 1 to 4 disclose techniques for detecting such special singing techniques and evaluating singing.

Prior Art Documents

Patent Documents

[0003]

Patent Document 1

Patent Document 2

Patent Document 3

Patent Document 4

Summary of the Invention

Problems to be Solved by the Invention

[0004] An object of the present invention is to provide a karaoke device capable of determining whether whisper voice singing (described later) is included in karaoke singing.

Means for Solving the Problems

[0005] Whisper voice singing is a singing technique that produces a voice that sounds like a whisper by reducing the vibration of the vocal cords relative to the amount of exhaled air. The inventors of this invention analyzed the spectral contrast based on singing voice signals that are perceived as whisper voice singing and found that the spectral contrast of whisper voice singing is smaller than that of normal singing in multiple frequency bands. This invention was completed based on this discovery and is a technology that can determine whisper voice singing by detecting the unique spectral contrast characteristics of this singing style.

[0006] Specifically, the invention for achieving the above objective is a karaoke device comprising: an acquisition unit that acquires singing voice signals associated with a singer's karaoke singing at predetermined intervals; and a determination unit that determines whether the singer's karaoke singing is whisper voice singing based on the spectral contrast obtained from the singing voice signals at predetermined intervals acquired by the acquisition unit. Other features of the present invention will be revealed in the specification and drawings described below. [Effects of the Invention]

[0007] According to the present invention, it is possible to determine whether or not whisper singing is included in karaoke singing. [Brief explanation of the drawing]

[0008] [Figure 1] This figure shows an example of the hardware configuration of a karaoke device according to an embodiment. [Figure 2A] This figure shows an example of the frequency spectrum of normal singing according to the embodiment. [Figure 2B] This figure shows an example of the frequency spectrum of whisper voice singing according to the embodiment. [Figure 3A] This figure shows an example of spectral contrast in normal singing according to the embodiment. [Figure 3B]This figure shows an example of spectral contrast in whisper voice singing according to the embodiment. [Figure 4] This figure shows an example of the software configuration of a karaoke unit according to the embodiment. [Figure 5] This is a flowchart showing the processing performed by the karaoke device according to the embodiment. [Modes for carrying out the invention]

[0009] <Embodiment> A karaoke device 1 according to an embodiment will be described with reference to Figures 1 to 5.

[0010] ==Karaoke Equipment== Karaoke device 1 is a device for karaoke performance and for singers to sing karaoke. Karaoke device 1 registers the karaoke songs selected by the singer in a reservation queue and performs the karaoke in order. As shown in Figure 1, karaoke device 1 comprises a karaoke unit 10, speakers 20, a display device 30, a microphone 40, and a remote control device 50.

[0011] The speaker 20 is configured to emit sound based on the sound signal emitted from the karaoke unit 10. The display device 30 is configured to display video and images on the screen based on the signal from the karaoke unit 10. The microphone 40 is configured to convert the singer's voice (sound input to the microphone 40) into an analog singing voice signal and input it to the karaoke unit 10. The remote control device 50 is a device for performing various operations on the karaoke unit 10. The singer can use the remote control device 50 to select the karaoke song they wish to sing. Icons for inputting instructions for various operations are displayed on the display screen of the remote control device 50.

[0012] The karaoke main unit 10 performs various controls related to karaoke singing, such as performance control of the selected karaoke music, display control of lyrics, background images, etc., and processing of the singing voice signal input through the microphone 40. As shown in FIG. 1, the karaoke main unit 10 includes a control unit 11, a communication unit 12, a storage unit 13, an audio processing unit 14, a display processing unit 15, and an operation unit 16. Each component is connected to the bus B via an interface (not shown).

[0013] The control unit 11 includes a CPU 11a and a memory 11b. The CPU 11a realizes various control functions by executing the programs stored in the memory 11b. The memory 11b is a storage device that stores the programs to be executed by the CPU 11a and temporarily stores various information during program execution.

[0014] The communication unit 12 provides an interface for connecting the karaoke main unit 10 to a communication line via a router (not shown).

[0015] The storage unit 13 is a large-capacity storage device that stores various data, such as a hard disk drive. The storage unit 13 stores a plurality of music data for performing karaoke performances by the karaoke device 1.

[0016] The music data is assigned a music ID for identifying each individual karaoke music. The music data includes accompaniment data, reference data, etc. The accompaniment data is the data that serves as the source of the karaoke performance sound. The accompaniment data includes information indicating the tempo during karaoke performance. The tempo is set to a predetermined value for each music. The reference data is the data used as a reference when scoring the karaoke singing by the singer.

[0017] In addition, the storage unit 13 stores lyric telop data for causing a display device 30 or the like to display lyrics corresponding to each karaoke song, background image data such as background images to be displayed on the display device 30 or the like during karaoke performance, performance time data indicating the karaoke performance time for each song, and attribute information of the song (information related to the song such as the name of the singer, the names of the lyricist and composer, and the genre).

[0018] Here, the storage unit 13 according to the present embodiment stores a learned model.

[0019] The learned model according to the present embodiment outputs a determination result as to whether the karaoke singing based on the singing voice signal is whisper voice singing.

[0020] The learned model is constructed by learning, as teacher data, data in which spectral contrasts for each of a plurality of frequency bands obtained from samples of the singing voice signal are associated with a singing method (normal singing or whisper voice singing). Note that normal singing is karaoke singing using chest voice.

[0021] The spectral contrast can be obtained by the following mathematical formula (1).

[0022] [Number] C is the spectral contrast, P is the band peak, and V is the band valley.

[0023] Specifically, the spectral contrast can be obtained by performing a discrete Fourier transform on the singing voice signal and taking the difference between the band peak and the band valley in a series X={X0, X1, X2, …, XN-1} obtained by cutting out the result (frequency spectrum) for each frequency band. FIG. 2A is an example of the frequency spectrum of normal singing, and FIG. 2B is an example of the frequency spectrum of whisper voice singing. In each figure, the horizontal axis represents frequency and the vertical axis represents amplitude.

[0024] The band peaks P and band valleys V can be calculated using the sequence X'={X'0,X'1,X'2,…,X'N-1} obtained by rearranging the sequence X in descending order, and the following formulas:

[0025]

number

[0026]

number

[0027] Figure 3A shows an example of spectral contrast in normal singing, and Figure 3B shows an example of spectral contrast in whisper singing. In each figure, the horizontal axis represents frequency, and the vertical axis represents the magnitude (power) of spectral contrast. In the waveforms in Figures 3A and 3B, the peaks correspond to band peaks, and the troughs correspond to band valleys. The frames in each figure indicate a range of a certain frequency band. As is clear from these figures, in a certain frequency band, the spectral contrast of whisper singing is smaller than that of normal singing.

[0028] The singing audio signal samples used to build the trained model will be selected from karaoke performances that differ in song genre, singing style, singer's age, gender, etc. Furthermore, to avoid bias, singing audio signals from multiple singers performing the same song will be used.

[0029] Furthermore, the singing method included in the training data (normal singing or whisper singing) is determined by listening to samples of singing audio signals from which spectral contrast has been calculated.

[0030] The server device (not shown) constructs a pre-trained model using training data. In this example, the server device (not shown) performs deep learning to learn features in spectral contrast, thereby constructing a neural network as a pre-trained model that takes spectral contrast as input and outputs singing methods.

[0031] The feature vector is the magnitude of the spectral contrast for each frequency band. A neural network, such as a CNN (Convolutional Neural Network), has an input layer that accepts spectral contrast input, an output layer that outputs singing methods, and a hidden layer that extracts feature vectors. The input layer passes the input spectral contrast to the hidden layer. The hidden layer passes the extracted feature vectors to the output layer. The output layer identifies the singing method based on the feature vectors output from the hidden layer.

[0032] Furthermore, the server device (not shown) inputs spectral contrast to the input layer and obtains the singing method results from the output layer. The singing method results output from the output layer are compared with the ground truth data, and the parameters are optimized so that the singing method results approach the ground truth data. Parameters include, for example, the weights (connection coefficients) between neurons and the coefficients of the activation function used in each neuron. The method of parameter optimization is not particularly limited, but for example, various parameters can be optimized using backpropagation.

[0033] The server device (not shown) transmits the constructed trained model to the karaoke device. The storage unit 13 of the karaoke device 1 stores the received trained model.

[0034] Furthermore, the trained model may be constructed using other learning algorithms, not limited to neural networks, such as SVM (Support Vector Machine), Bayesian networks, regression trees, and random forests.

[0035] The sound processing unit 14 controls the performance of karaoke songs and processes the singing voice signals input through the microphone 40, based on the control of the control unit 11. The display processing unit 15 processes various displays on the display device 30 and the remote control device 50, based on the control of the control unit 11. For example, the display processing unit 15 controls the display device 30 to display an image on which lyrics and various icons are superimposed on the background image during the performance of a karaoke song. Alternatively, the display processing unit 15 displays various icons for operation input on the display screen of the remote control device 50. The operation unit 16 consists of a panel switch and a remote control receiving circuit, and outputs operation signals such as song selection signals and performance stop signals to the control unit 11 in response to the singer's operation of the panel switch of the karaoke device 1 or the remote control device 50. The control unit 11 detects the operation signals from the operation unit 16 and executes the corresponding processing.

[0036] (Software configuration) Figure 3 shows an example of the software configuration of the karaoke unit 10 according to this embodiment. The karaoke unit 10 comprises an acquisition unit 100, a determination unit 200, and a presentation unit 300. The acquisition unit 100, the determination unit 200, and the presentation unit 300 are realized by the CPU 11a executing a program stored in memory 11b.

[0037] [Acquisition Department] The acquisition unit 100 acquires the singing audio signal associated with the singer's karaoke performance at predetermined intervals.

[0038] A predetermined interval is set in advance, for example, as one frame (2048 samples).

[0039] A singer using the karaoke machine 1 operates the remote control device 50 to select the song they wish to sing. The karaoke machine 1 plays the karaoke version of the song selected by the singer. The singer sings along to the karaoke version. The acquisition unit 100 acquires the singing voice signal input through the microphone 40 at predetermined intervals in conjunction with the karaoke singing. The acquisition of the singing voice signal may be performed sequentially as the karaoke singing of the song progresses, or it may be performed all at once after the karaoke singing of the song has been completed.

[0040] [Judgment section] The determination unit 200 determines whether the singer's karaoke singing is whisper singing based on the spectral contrast obtained from the singing audio signal for each predetermined section acquired by the acquisition unit 100.

[0041] The judgment unit 200 analyzes the singing voice signal and divides it into multiple frequency bands with predetermined widths. The more frequency bands there are, the more accurate the judgment becomes. On the other hand, as the amount of data increases, the number of elements increases, and it takes time to build a trained model. Therefore, it is desirable to set an appropriate number of frequency bands. For example, the number of frequency bands is set to one, such as 9. The width of the frequency bands is also set to one, such as in increments of 1.0 kHz. It is also possible that some frequency values ​​overlap between one frequency band and other frequency bands.

[0042] The determination unit 200 calculates the spectral contrast for each divided frequency band. The spectral contrast can be calculated using the method described above.

[0043] Furthermore, the difference in spectral contrast between normal singing and whisper singing tends to be more pronounced in the frequency range above approximately 2 kHz. Therefore, the determination unit 200 may divide the frequency range above approximately 2 kHz into multiple frequency bands with predetermined widths.

[0044] The determination unit 200 determines, based on the calculated spectral contrast, whether the singer's karaoke performance is a whisper voice performance.

[0045] In this embodiment, the determination unit 200, upon receiving spectral contrast obtained from singing audio signals for predetermined intervals acquired by the acquisition unit 100, uses a pre-trained model that has been trained to output a determination result on whether the karaoke singing based on the singing audio signal is whisper singing to determine whether the singer's karaoke singing is whisper singing.

[0046] Specifically, the determination unit 200 inputs the calculated spectral contrast for each frequency band (i.e., spectral contrast in a predetermined interval) to the trained model stored in the memory unit 13. The trained model determines whether the input spectral contrast is for normal singing or for whisper singing, and outputs a determination result indicating that it is either normal singing or whisper singing.

[0047] If the trained model outputs a result indicating whispery singing, the determination unit 200 determines that the karaoke singing in the predetermined section is whispery singing. On the other hand, if the trained model outputs a result indicating normal singing, the determination unit 200 determines that the karaoke singing in the predetermined section is normal singing. The determination unit 200 outputs the determination result to the presentation unit 300.

[0048] [Presentation part] If the display unit 300 determines that a singer's karaoke performance is a whisper voice performance, it displays the determination result or information based on the determination result to the singer.

[0049] For example, suppose whisper singing takes place in a predetermined section. In this case, the display unit 300 can display an icon on the display device 30 indicating that whisper singing has taken place. Alternatively, the display unit 300 may notify the singer of the fact that whisper singing has taken place via sound (applause, cheers, etc.) through the speaker 20. These are examples of "presenting the judgment result to the singer."

[0050] Alternatively, when the display unit 300 displays the scoring result for karaoke singing, it can display a scoring result that includes points added according to the number of predetermined sections in which whisper voice singing was performed. This is an example of "presenting information based on the judgment result to the singer."

[0051] ==Regarding the operation of karaoke machine 1== Next, a specific example of the operation of the karaoke device 1 in this embodiment will be described with reference to Figure 5. Figure 5 is a flowchart showing an example of the operation of the karaoke device 1. In this example, the memory unit 13 stores a trained model that, when inputting spectral contrast obtained from the singing voice signal, outputs a determination result of whether the karaoke singing based on the singing voice signal is whisper voice singing. In this example, it is assumed that the determination of whether or not it is whisper voice singing is made at predetermined intervals.

[0052] The singer operates the remote control device 50 to select the song they wish to sing karaoke to. The karaoke device 1 starts playing the selected song (karaoke performance starts. Step 10).

[0053] The acquisition unit 100 acquires the singing audio signal associated with the singer's karaoke performance at predetermined intervals (acquisition of singing audio signal at predetermined intervals; step 11).

[0054] The determination unit 200 receives the spectral contrast obtained from the singing audio signal acquired in step 11 as input to the trained model. The trained model outputs a determination result indicating whether the karaoke singing based on the singing audio signal is whisper singing. Based on the output determination result, the determination unit 200 determines whether the singer's karaoke singing is whisper singing (determination of whether it is whisper singing; step 12).

[0055] If the singer's karaoke performance is determined to be a whisper voice performance (the result is Y in step 13), the presentation unit 300 presents the determination result to the singer (presentation of determination result; step 14).

[0056] The karaoke device 1 repeatedly performs the processes from step 11 to step 14 until the karaoke performance of the song is finished (if the result is Y in step 15).

[0057] Specifically, singer S operates the remote control device 50 to select the song M that they wish to sing karaoke to. The karaoke device 1 starts playing the selected song M. Singer S begins singing karaoke along with the karaoke music.

[0058] The acquisition unit 100 acquires the singing audio signal for the first frame (2048 samples) as the singer S sings karaoke. The acquisition unit 100 outputs the acquired singing audio signal to the determination unit 200.

[0059] The determination unit 200 performs a discrete Fourier transform on the singing audio signal output from the acquisition unit 100 to obtain a discrete spectrum. The determination unit 200 divides the obtained discrete spectrum into nine frequency bands (F1 to F9).

[0060] In this example, the frequency bandwidths are assumed to be 1.5kHz to 2.5kHz, 1.5kHz to 4.0kHz, 3.0kHz to 4.0kHz, 4.0kHz to 5.0kHz, 5.0kHz to 6.0kHz, 5.0kHz to 7.0kHz, 6.0kHz to 7.0kHz, 7.0kHz to 8.0kHz, and 8kHz to 9kHz. Furthermore, in this example, the sampling frequency is assumed to be 44100Hz.

[0061] In this case, the determination unit 200 uses the Discrete Fourier Transform to determine the amplitude at 2048 frequency points approximately every 21.533 Hz. That is, the determination unit 200 determines the amplitude X1 at frequency 0, the amplitude X2 at frequency 21.533 Hz, the amplitude X3 at frequency 43.066 Hz, the amplitude X4 at frequency 64.599 Hz, the amplitude X5 at frequency 86.132 Hz, the amplitude X6 at frequency 107.666 Hz, ... and the amplitude X2048 at frequency 44078.466 Hz (a total of 2048 points).

[0062] The determination unit 200 takes a list of the calculated amplitudes, extracting them for each frequency band, and defines this list as sequence Xn. For example, since the number of amplitudes for frequencies included in frequency band F1 (1.5kHz~2.5kHz) is 47, the sequence X1 = {X1, X2, ..., X47}.

[0063] The determination unit 200 generates a sequence X1' by rearranging the sequence X, which is arranged in frequency order, in order of decreasing amplitude value.

[0064] Based on formula 2, the determination unit 200 calculates the value of the band peak P1 (average amplitude) by adding the amplitudes included in a predetermined fixed percentage (for example, 11 if the percentage is 25%) from the amplitudes included in the series X1', in descending order of amplitude value, and then dividing by the number of added amplitudes (11 in the above example).

[0065] Similarly, the determination unit 200 calculates the value of the band valley V1 (average amplitude) by adding the amplitudes included in a predetermined fixed percentage (for example, 11 if the percentage is 25%) from the amplitudes included in the series X1', in order from smallest to largest, based on formula 3, and then dividing by the number of added amplitudes (11 in the above example).

[0066] The determination unit 200 determines the spectral contrast C1 in the frequency band F1 by applying the calculated band peak P1 value and band valley V1 value to equation 1.

[0067] The determination unit 200 performs the same processing from frequency band F2 to frequency band F9 to determine the spectral contrast C2 in frequency band F2 and the spectral contrast C9 in frequency band F9.

[0068] The determination unit 200 inputs the spectral contrasts C1 to C9 obtained from the determination unit 13 to the trained model stored in the memory unit 13. The trained model outputs a determination result indicating whether the input spectral contrasts indicate whisper voice singing. Based on the output determination result, the determination unit 200 determines whether the singer's karaoke singing is whisper voice singing.

[0069] For example, if the trained model outputs a result indicating that spectral contrast C1 to spectral contrast C9 indicates whisper voice singing, the determination unit 200 determines that singer S's karaoke performance is whisper voice singing. In this case, singer S is performing whisper voice singing in the first frame of the karaoke performance of song M.

[0070] If it is determined that singer S's karaoke performance is a whispery voice performance, the presentation unit 300 presents the determination result to singer S.

[0071] Karaoke device 1 performs the above process every frame until the karaoke performance of song M is finished (until singer S finishes singing karaoke), determining whether or not it is whisper singing and displaying the determination result.

[0072] As is clear from the above, the karaoke device 1 according to this embodiment includes an acquisition unit 100 that acquires singing voice signals associated with a singer's karaoke singing at predetermined intervals, and a determination unit 200 that determines whether the singer's karaoke singing is whisper voice singing based on the spectral contrast obtained from the singing voice signals at predetermined intervals acquired by the acquisition unit 100.

[0073] More specifically, the determination unit 200, upon receiving spectral contrast obtained from singing audio signals for predetermined intervals acquired by the acquisition unit 100, uses a pre-trained model that has been trained to output a determination result on whether the karaoke singing based on the singing audio signal is whisper singing to determine whether the singer's karaoke singing is whisper singing.

[0074] In other words, the karaoke device 1 according to this embodiment can use a trained model to determine whether or not whisper singing is included in the karaoke performance.

[0075] Furthermore, the karaoke device 1 according to this embodiment has a display unit 300 that, when it is determined that the singer's karaoke singing is whisper voice singing, presents the determination result or information based on the determination result to the singer. With such a karaoke device 1, the result that the singer has been determined to be whisper voice singing can be presented to the singer directly or indirectly.

[0076] <Example 1> The output of a trained model is not limited to the examples of the embodiment. For example, a trained model can be used to output trend data indicating whether or not karaoke singing based on a singing voice signal is whisper singing, when spectral contrast obtained from a singing voice signal is input.

[0077] Trend data indicates the tendency of whether or not a song is sung in a whispery voice. For example, trend data can be represented by a value of "0" for normal singing and "1" for whispery voice singing.

[0078] The determination unit 200 inputs the calculated spectral contrast to the trained model stored in the memory unit 13. The trained model outputs trend data indicating whether the input spectral contrast is whisper voice singing. Based on the output trend data, the determination unit 200 determines whether the singer's karaoke performance is whisper voice singing.

[0079] Whether or not singing is done in a whisper voice can be determined based on a pre-set threshold. For example, as described above, if the trend data is shown as a value between 0 and 1, the determination unit 200 checks whether the output trend data value is equal to or greater than the pre-set threshold of 0.5. If the trend data value is equal to or greater than 0.5, the determination unit 200 determines that the singer's karaoke performance is in a whisper voice.

[0080] As is clear from the above, the determination unit 200 in the karaoke device 1 according to this modified example can determine whether a singer's karaoke performance is whispery when it receives spectral contrast data obtained from singing audio signals for predetermined intervals acquired by the acquisition unit 100, using a pre-trained model that has been trained to output trend data indicating whether the karaoke performance based on the singing audio signal is whispery. In other words, the karaoke device 1 according to this modified example can determine whether a karaoke performance includes whispery singing using a pre-trained model.

[0081] <Modification 2> It is also possible to determine whether a singer's karaoke performance is a whispery vocal performance without using a pre-trained model.

[0082] In this modified example, the memory unit 13 stores a preset threshold value for determining whether the spectral contrast for each frequency band is that of whisper voice singing. The threshold value can be, for example, an intermediate value between the spectral contrast of normal singing and the spectral contrast of whisper voice singing for each frequency band.

[0083] Similar to the embodiment, the determination unit 200 checks whether the obtained spectral contrast value satisfies predetermined conditions. If the predetermined conditions are met, the determination unit 200 determines that the singer's karaoke singing is whisper voice singing. The predetermined conditions are conditions for determining that the singer's karaoke singing is whisper voice singing, such as "more than half of the spectral contrast values ​​in a predetermined interval are below a threshold."

[0084] For example, similar to the embodiment, the determination unit 200 determines the spectral contrast C9 in frequency band F9 from the spectral contrast C1 in frequency band F1.

[0085] The determination unit 200 checks whether the obtained spectral contrasts C1 to C9 satisfy a predetermined condition. For example, suppose the predetermined condition is "at least half of the spectral contrast values ​​in a predetermined interval are below the threshold." In this case, the determination unit 200 determines the number of spectral contrasts that are below the threshold. The determination unit 200 checks whether the number of obtained spectral contrasts is at least half of the total number of spectral contrasts (at least 5 in this example). If the predetermined condition is met, the determination unit 200 determines that the singer S's karaoke singing is whisper voice singing.

[0086] As is clear from the above, the determination unit 200 in the karaoke device 1 according to this modified example uses a predetermined threshold for determining whether the spectral contrast for each frequency band is whisper voice singing, and can determine whether the singer's karaoke singing is whisper voice singing based on the spectral contrast obtained from the singing audio signal for each predetermined section acquired by the acquisition unit 100. In other words, the karaoke device 1 according to this modified example can determine whether whisper voice singing is included in the karaoke singing without using a trained model.

[0087] <Other> It is also possible to supply the program to a computer using a non-transitory computer-readable medium (with an executable program thereon) on which the above program is stored. Examples of non-transitory computer-readable media include magnetic recording media (e.g., flexible disks, magnetic tapes, hard disk drives), CD-ROMs (Read Only Memory), etc.

[0088] The above embodiments are presented as examples and do not limit the scope of the invention. The above configurations can be combined as appropriate, and various omissions, substitutions, and modifications can be made without departing from the spirit of the invention. The above embodiments and their variations are included in the scope and spirit of the invention, as well as in the claims of the invention and its equivalents. [Explanation of symbols]

[0089] 1. Karaoke machine 10 Karaoke machine 100 Acquisition Department 200 Judgment section 300 Presentation section

Claims

1. An acquisition unit that acquires the singing audio signal associated with the singer's karaoke performance at predetermined intervals, A determination unit determines whether the singer's karaoke singing is whisper voice singing based on the spectral contrast obtained from the singing audio signal for each predetermined section acquired by the acquisition unit, A karaoke device having the following features.

2. The karaoke apparatus according to claim 1, characterized in that the determination unit, upon input of spectral contrast obtained from singing audio signals for predetermined intervals acquired by the acquisition unit, determines whether the karaoke singing based on the singing audio signals is whisper voice singing using a pre-trained model that has been trained to output a determination result of whether or not the karaoke singing based on the singing audio signals is whisper voice singing.

3. The karaoke apparatus according to claim 1, characterized in that the determination unit, upon input of spectral contrast obtained from singing audio signals for predetermined intervals acquired by the acquisition unit, determines whether the singer's karaoke singing is whisper voice singing using a pre-trained model that has been trained to output trend data indicating whether the karaoke singing based on the singing audio signal is whisper voice singing.

4. The karaoke apparatus according to claim 1, characterized in that the determination unit uses a predetermined threshold for determining whether the spectral contrast for each frequency band is whisper voice singing, and determines whether the singer's karaoke singing is whisper voice singing based on the spectral contrast obtained from the singing audio signal for each predetermined section acquired by the acquisition unit.

5. The karaoke device according to any one of claims 1 to 4, characterized in that, if the karaoke singing of the singer is determined to be whisper voice singing, the device has a display unit that presents the determination result or information based on the determination result to the singer.

Images