JP6670696B2 - Karaoke equipment - Google Patents

Description

  The present invention relates to a karaoke apparatus.

  In karaoke singing, a singing voice of a singer input from a microphone is amplified by an amplifier and output (sound emission) from a speaker. Here, in an environment such as a karaoke box where a speaker and a microphone are in the same space, howling occurs when the microphone picks up a singing voice or a karaoke performance sound output from the speaker.

  Here, there is a method of suppressing howling using a notch filter. For example, in Patent Document 1, when a difference between a peak frequency extracted from a singing signal and a target frequency based on guide melody data (reference data) exceeds a predetermined value (for example, 200 cents), the peak frequency corresponds to howling. A technique that suppresses howling by determining that the frequency is a peak frequency and changing the filter characteristic so that the notch frequency becomes the peak frequency is disclosed (see paragraphs [0009] to [0011] in Patent Document 1). 1 to 4).

Patent No. 4241296

  By the way, a case where the frequency corresponding to howling and the target frequency are approximated may be considered. In this case, according to the technique of Patent Document 1, it is determined that the difference between the peak frequency and the target frequency does not exceed a predetermined value. Therefore, although the peak frequency is actually a frequency corresponding to howling, it is not possible to filter a signal of the frequency. That is, when the frequency corresponding to the howling and the target frequency are close to each other, it is difficult to suppress the howling with the technique of Patent Document 1.

  On the other hand, when the frequency corresponding to howling and the target frequency are close to each other, it is conceivable to use a notch filter for filtering a signal having a frequency corresponding to howling. However, such a notch filter also filters a signal of an approximate target frequency. That is, when the singing voice signal includes the target frequency, the singing voice signal is also filtered. As a result, the output level of the singing voice signal decreases, and the singing voice becomes difficult to hear. This becomes more remarkable as the song is sung along the reference data (that is, as the song is sung well). Therefore, comfortable karaoke songs cannot be performed.

  An object of the present invention is to provide a karaoke apparatus capable of suppressing both howling and a decrease in the output level of a singing voice signal when a frequency corresponding to howling and a target frequency based on reference data are close to each other. It is in.

A main invention for achieving the above object is a karaoke apparatus having a notch filter for filtering a signal having a frequency corresponding to howling, based on an audio signal input from a microphone along with karaoke singing of a music piece. A howling detection unit that detects howling, a frequency corresponding to the detected howling, a determination unit that determines whether or not a target frequency based on the reference data of the music is approximated, and a frequency corresponding to the howling. A karaoke apparatus comprising: a filter setting unit that sets the notch filter with a notch depth smaller than the predetermined notch depth when it is determined that the target frequency is close to the target frequency.
Other features of the present invention will be apparent from the description in the specification and the drawings described below.

  According to the present invention, when the frequency corresponding to howling and the target frequency based on the reference data are close to each other, it is possible to suppress both howling and a decrease in the output level of the singing voice signal.

It is a figure showing the outline of the karaoke device concerning a 1st embodiment. It is a figure showing the example of hardware constitutions of the karaoke main part concerning a 1st embodiment. FIG. 3 is a diagram illustrating an example of a hardware configuration of a sound processing unit according to the first embodiment. FIG. 3 is a diagram illustrating an example of a software configuration of a karaoke body according to the first embodiment. FIG. 4 is a diagram illustrating filter characteristics of the notch filter according to the first embodiment. FIG. 4 is a diagram illustrating filter characteristics of the notch filter according to the first embodiment. FIG. 4 is a diagram illustrating filter characteristics of the notch filter according to the first embodiment. 5 is a flowchart illustrating an operation example of the karaoke device according to the first embodiment. It is a figure showing the example of software constitutions of the karaoke body concerning a 2nd embodiment.

  From the description of the specification and the drawings described below, at least the following matters will be apparent in addition to the above main invention.

  That is, when the filter setting unit determines that the frequency corresponding to the howling and the target frequency are not approximate, a karaoke apparatus that sets the notch filter at a predetermined notch depth set in advance is apparent. . According to such a karaoke apparatus, howling can be suppressed without affecting the output level of the singing voice signal.

  Further, a storage unit that stores the degree of deviation of the singing voice with respect to the reference data as history data for each singer, wherein the determination unit corrects the target frequency based on the history data of the singer performing the karaoke singing. A karaoke apparatus that determines whether or not the corrected target frequency and the frequency corresponding to the howling are close to each other will be apparent. According to such a karaoke apparatus, it is possible to more accurately determine whether the target frequency and the frequency corresponding to howling are close to each other.

<First embodiment>
The karaoke apparatus 1 according to the first embodiment will be described with reference to FIGS.

== Karaoke device ==
(Hardware configuration)
The karaoke apparatus 1 is an apparatus for performing karaoke performance of a song selected by a singer and singing a karaoke song by the singer. As shown in FIG. 1, the karaoke apparatus 1 includes a karaoke body 10, a speaker 20, a display device 30, a microphone 40, and a remote control device 50.

  The karaoke main body 10 performs various controls related to karaoke singing, such as control of performance of a pre-registered tune, display control of lyrics and a background image, and processing of an audio signal input through the microphone 40. FIG. 2 is a diagram illustrating an example of a hardware configuration of the karaoke body 10. The karaoke main body 10 includes a control unit 11, a communication unit 12, a storage unit 13, a sound 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).

  The control unit 11 includes a CPU 11a and a memory 11b. The CPU 11a realizes various control functions by executing an operation program stored in the memory 11b. The memory 11b is a storage device that stores a program to be executed by the CPU 11a and temporarily stores various information when the program is executed.

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

  The storage unit 13 is a large-capacity storage device that stores various data. The storage unit 13 stores music data for performing a karaoke performance by the karaoke device 1. The music data is provided with a predetermined ID (music ID) for each music. The music data includes MIDI data, reference data, and the like. The reference data is composed of time-series information indicating the pitch and the duration of each note constituting the music. The reference data is used, for example, as data that is used as a reference when singing a karaoke song by a singer. The storage unit 13 also stores lyrics telop data for displaying the lyrics telop corresponding to the karaoke tune on the display device 30 and the like, and background image data such as a background image displayed on the display device 30 and the like during the karaoke performance. .

  The sound processing unit 14 controls the performance of the karaoke music and processes the audio signal input through the microphone 40 based on the control of the control unit 11.

  As shown in FIG. 3, the acoustic processing unit 14 according to the present embodiment includes a MIDI sound source 140, a notch filter 141, a mixer 142, and an amplifier 143. The MIDI sound source 140 generates a tone signal based on music data. The notch filter 141 is a filter for filtering a signal having a frequency corresponding to howling from an audio signal input from the microphone 40. The setting of the notch filter 141 will be described later. The mixer 142 mixes the tone signal from the MIDI sound source 140 and the audio signal from the notch filter 141 at an appropriate ratio, and outputs the resulting signal to the amplifier 143. The amplifier 143 amplifies the mixing signal from the mixer 142 with an amplifier circuit and outputs the amplified signal to the speaker 20 as a sound emission signal. With such a configuration, a karaoke performance sound based on a tone signal and a singing voice based on a voice signal from a microphone 40 are emitted from the speaker 20.

  The display processing unit 15 performs processing related to various displays on the display device 30 based on the control of the control unit 11. For example, the display processing unit 15 causes the display device 30 to display a video in which lyrics telops and various icons are superimposed on a background video during a karaoke performance.

  The operation unit 16 includes a panel switch, a remote control receiving circuit, and the like, and sends an operation signal, such as a reservation signal and a performance stop signal, to the control unit 11 in response to an operation of the singer by operating the panel switch of the karaoke device 1 or the remote control device 50. Output. The control unit 11 detects an operation signal from the operation unit 16 and executes a corresponding process.

  As described above, the speaker 20 is configured to emit sound based on a sound emission signal from the karaoke body 10. The display device 30 is configured to display a video or an image on a screen based on a signal from the karaoke body 10. The microphone 40 is configured to convert an input singer's voice (singing voice) or the like into an analog voice signal and input the analog voice signal to the karaoke body 10.

  The remote control device 50 is a device for performing various operations on the karaoke body 10. The singer can use the remote control device 50 to make reservations for music and the like. The display of the remote controller 50 displays a GUI for inputting various operation instructions.

(Software configuration of the karaoke unit)
FIG. 4 is a diagram illustrating an example of a software configuration of the karaoke body 10. The karaoke body 10 includes a howling detection unit 100, a determination unit 200, and a filter setting unit 300. The howling detection section 100, the determination section 200, and the filter setting section 300 are realized by the CPU 11a executing a program stored in the memory 11b.

[Howling detector]
The howling detection unit 100 detects howling based on an audio signal input from the microphone 40 along with singing a karaoke song.

  As described above, howling may occur when the microphone 40 picks up the singing voice or the karaoke performance sound output from the speaker 20. In this case, the audio signal input from the microphone 40 includes a frequency corresponding to howling in addition to a frequency corresponding to the singing voice of the singer. Therefore, howling detecting section 100 detects a frequency corresponding to howling included in the audio signal.

  As a method for detecting howling included in the audio signal, a known method can be used. For example, the howling detection unit 100 divides the audio signal into a plurality of frequency bands, compares the peak level of the audio signal for each of the divided frequency bands with a threshold of a predetermined level, and determines that the peak level of the audio signal has If the value is exceeded, it is determined that howling has occurred, and howling is detected.

[Judgment unit]
The determining unit 200 determines whether or not the frequency corresponding to the detected howling is similar to the target frequency based on the reference data of the music.

  The target frequency based on the music reference data is obtained by converting pitch data included in the music reference data into a frequency. Since the reference data is information indicating the pitch and the duration of each note constituting the music, the singer generally sings along the reference data. That is, the frequency converted from the pitch data included in the reference data corresponds to the pitch targeted by the singer.

  When a howling is detected by the howling detection unit 100, the determination unit 200 reads out the pitch data of the music corresponding to the time when the audio signal is input from the storage unit 13, performs frequency conversion, and obtains a target frequency. . Then, determination section 200 obtains a difference between the frequency corresponding to the detected howling and the target frequency. When the difference is equal to or smaller than the predetermined value, the determination unit 200 determines that the frequency corresponding to the howling is close to the target frequency. When the difference is larger than the predetermined value, the determination unit 200 determines that the frequency corresponding to the howling and the target frequency are not approximated. The predetermined value is a pitch value serving as a reference for determining the approximation of two different frequencies. The predetermined value is set, for example, to 200 cents. Note that “approximation” includes that a frequency corresponding to howling matches a target frequency.

[Filter setting section]
When it is determined that the frequency corresponding to the howling and the target frequency are not approximate, the filter setting unit 300 sets the notch filter at a predetermined notch depth set in advance. On the other hand, when it is determined that the frequency corresponding to howling and the target frequency are approximate, filter setting section 300 sets the notch filter with a notch depth smaller than a predetermined notch depth.

  The notch depth is a value indicating the degree (level) when the notch filter 141 filters a signal in a frequency band corresponding to howling. For example, when the notch depth D1 is set and the degree of filtering is higher than the notch depth D1 (when more signals are filtered), the notch depth is referred to as "deep". When the degree of filtering is lower than the depth D1 (when more signals are transmitted), the notch depth is referred to as “shallow”.

  5A to 5C are diagrams illustrating the filter characteristics of the notch filter 141. The horizontal axis is the frequency, and the vertical axis is the level of the notch depth. FIG. 5A shows the filter characteristics of the notch filter 141 when howling is not detected. FIG. 5B is a diagram illustrating the filter characteristics of the notch filter 141 when it is determined that the frequency f1 corresponding to howling and the target frequency f2 are not approximated. FIG. 5C is a diagram illustrating filter characteristics of the notch filter 141 when it is determined that the frequency f1 corresponding to howling matches the target frequency f2.

  As apparent from FIG. 5A, when the howling is not detected, the level of the notch filter 141 is 0. In this case, since the notch filter 141 transmits the audio signal of any frequency band at the same level, the audio signal input from the microphone 40 is mixed with the musical tone signal as it is and output to the amplifier 143.

  On the other hand, when the howling occurs, it is necessary to change the filter characteristic of the notch filter 141 to suppress the howling.

  When howling is detected by howling detecting section 100, determining section 200 determines whether or not frequency f1 corresponding to howling is approximate to target frequency f2.

  When it is determined that the frequency f1 corresponding to howling and the target frequency f2 are not approximate, the frequencies are separated as shown in FIG. 5B. Therefore, even if a filter setting for filtering a signal having a frequency corresponding to howling is performed, it is unlikely that the singing voice signal obtained by the singing will be filtered as long as the singing is performed according to the reference data.

  Therefore, the filter setting unit 300 sets the notch filter at a predetermined notch depth set in advance. For example, as shown in FIG. 5B, the filter setting unit 300 sets the notch depth of a portion corresponding to the frequency f1 to a predetermined notch depth D1 so as to filter a signal of the frequency f1 corresponding to howling. In this case, howling can be suppressed. Further, since the signal corresponding to the target frequency f2 is not filtered by the notch filter 141, the output level of the singing voice does not decrease.

  On the other hand, when it is determined that the frequency f1 corresponding to the howling matches the target frequency f2, when the signal of the frequency f1 corresponding to the howling is filtered, there is a high possibility that the singing voice signal is also filtered. In this case, the filtering of the singing voice signal becomes more remarkable as the singer performs the singing along the reference data.

  In this case, the filter setting unit 300 sets the notch filter 141 with a notch depth smaller than a predetermined notch depth. For example, as shown in FIG. 5C, the filter setting unit 300 filters the signal of the frequency f1 corresponding to the howling while transmitting the signal of the target frequency f2 to some extent, so that the notch depth shallower than the predetermined notch depth D1. The notch filter 141 is set so as to be D2. In this case, it is possible to prevent the output level of the singing voice signal from dropping significantly while suppressing howling to some extent.

  Note that how much the notch depth of the notch filter 141 is made smaller than the predetermined notch depth D1 can be appropriately set according to an input audio signal or the like. For example, a level at which a signal waveform specific to howling changes to a signal waveform indicating a human voice may be obtained, and that value may be used as the notch depth D2.

== Operation of Karaoke Device 1 ==
Next, a specific example of the operation of the karaoke apparatus 1 according to the present embodiment will be described with reference to FIG. FIG. 6 is a flowchart showing an operation example of the karaoke apparatus 1.

  The singer selects a music piece X for which karaoke singing is desired via the remote control device 50 (selection of music piece, step 10). The karaoke apparatus 1 plays the music piece X selected in step 10. The singer performs karaoke singing along with the performance (karaoke singing, step 11).

  The howling detection section 100 detects howling based on the audio signal input from the microphone 40 in association with the karaoke singing in step 11.

  When the howling is detected (in the case of Y in step 12), the determination unit 200 approximates the frequency f1 corresponding to the detected howling and the target frequency f2 based on the reference data of the music piece X selected in step 10. It is determined whether or not to do.

  When it is determined that the frequency f1 corresponding to howling and the target frequency f2 are not approximated (in the case of N in step 13), the filter setting unit 300 sets the notch filter 141 so that the notch depth D1 becomes a predetermined notch depth D1 set in advance. Is set (filter setting at a predetermined notch depth. Step 14).

  On the other hand, when it is determined that the frequency f1 corresponding to howling and the target frequency f2 are approximate (in the case of Y in step 13), the filter setting unit 300 has a notch depth D2 shallower than the predetermined notch depth D1. The notch filter 141 is set (filter setting with a shallow notch depth, step 15).

  As described above, according to the karaoke apparatus 1 according to the present embodiment, when the frequency corresponding to the howling and the target frequency based on the reference data are close to each other, both the howling and the decrease in the output level of the singing voice signal are suppressed. can do. Therefore, the singer and the audience are not discomforted by the howling. In addition, since the singing voice is also output to some extent, the more the user sings well, the more difficult it is to hear the singing voice, and the user can enjoy a comfortable karaoke singing.

<Second embodiment>
Next, a karaoke apparatus 1 according to a second embodiment will be described with reference to FIG. In the present embodiment, an example in which the target frequency is corrected based on the singer's history data will be described. A detailed description of the same configuration as in the first embodiment will be omitted.

  Karaoke singing is skillful by the singer. Therefore, some singers may always sing greatly from the reference data. When such a singer sings, even if the determination unit 200 determines that the frequency corresponding to howling and the target frequency are not approximate, the determination is not always appropriate. That is, since the singing voice of such a singer greatly deviates from the target frequency, the actual singing voice may be included in a frequency band similar to howling.

  In such a case, if the notch filter 141 is set at a predetermined notch depth, the signal of the singing voice is filtered, and the output level of the singing voice is reduced.

  Therefore, in the present embodiment, a configuration will be described in which it is determined whether or not the frequency corresponding to howling and the target frequency are close to each other in consideration of the singing characteristic of the singer.

  FIG. 7 is a diagram illustrating an example of a software configuration of the karaoke body 10. The karaoke body 10 includes a howling detection unit 100, a determination unit 200, a filter setting unit 300, and a history calculation unit 400. Howling detection section 100, determination section 200, filter setting section 300, and history calculation section 400 are realized by CPU 11a executing a program stored in memory 11b.

  The history calculation unit 400 compares the reference data of the song with the singing voice data obtained when the song is sung, and calculates the degree of deviation of the singing voice data from the reference data. In addition, the storage unit 13 according to the present embodiment stores the degree of deviation of the singing voice with respect to the reference data as history data for each singer.

  For example, it is assumed that singer X sings song A, song B, and song C. In this case, the history calculation unit 400 compares the reference data of the song A with the singing voice data obtained when singing the song A, and obtains a difference value (degree of deviation) of the singing voice data with respect to the reference data (pitch). . The difference value may be an average value of the entire music A, or a maximum value or a minimum value in the music A. The history calculation unit 400 outputs the obtained difference value to the storage unit 13 in association with the singer ID of the singer X. The history calculation unit 400 performs the same processing when singing the music B and the music C. The storage unit 13 stores the difference value as history data for each singer.

  Here, it is assumed that howling is detected when the singer X sings a new song D.

  In this case, the determination unit 200 corrects the target frequency based on the history data of the singer performing the karaoke singing, and determines whether or not the corrected target frequency and a frequency corresponding to howling are similar.

  Specifically, the determination unit 200 reads the history data (difference value F) of the singer X from the storage unit 13. Then, the determination unit 200 corrects the target frequency f2 to the target frequency f'2 by adding the difference value F included in the history data to the target frequency. The difference value F used at this time may be, for example, the maximum value among a plurality of difference values, or a value obtained by averaging the difference values of the music pieces A to C. The determination unit 200 obtains a difference between the corrected target frequency f'2 and the frequency f1 corresponding to howling, and determines whether the frequencies are close to each other.

  As described above, according to the karaoke apparatus 1 according to the present embodiment, by considering the singing characteristics of the singer, it is possible to more accurately determine whether the frequency corresponding to howling and the target frequency are close to each other. . Therefore, even a singer who has a large deviation from the reference data can enjoy a comfortable karaoke singing without significantly lowering the output level of the singing voice.

  The above embodiments and the like have been presented as examples and do not limit the scope of the invention. The above configurations can be appropriately combined and implemented, and various omissions, replacements, and changes can be made without departing from the spirit of the invention. The above-described embodiments and their modifications are included in the scope and gist of the invention, and are also included in the invention described in the claims and equivalents thereof.

DESCRIPTION OF SYMBOLS 1 Karaoke apparatus 10 Karaoke main body 100 Howling detection part 141 Notch filter 200 Judgment part 300 Filter setting part

Claims (3)

A karaoke apparatus having a notch filter for filtering a signal having a frequency corresponding to howling,
A howling detection unit that detects howling based on an audio signal input from a microphone along with the karaoke singing of the song,
A frequency corresponding to the detected howling, and a determination unit that determines whether a target frequency based on the reference data of the music piece is approximated,
If a frequency corresponding to the howling and the target frequency is determined to be approximate, a filter setting section configured to set the notch filter at a shallow notch depth than Jo Tokoro of notch depth,
Karaoke device having a karaoke device. Claim wherein the filter setting unit, when a frequency corresponding to the howling and the target frequency is determined not to be approximated, which is characterized in that setting the notch filter at the predetermined notch depth that is set in advance A karaoke apparatus according to claim 1. A storage unit that stores the degree of deviation of the singing voice with respect to the reference data as history data for each singer,
The determining unit corrects the target frequency based on history data of the singer performing the karaoke song, and determines whether the corrected target frequency and the frequency corresponding to the howling are close to each other. 3. The karaoke apparatus according to claim 1, wherein

Images