JP4134921B2 - Karaoke equipment - Google Patents

Description

  The present invention relates to a karaoke apparatus for scoring a singer's singing ability.

  Among karaoke apparatuses that perform automatic performance based on music data, there is a karaoke apparatus that inputs a singing voice signal and scores a singer's singing ability. This kind of karaoke apparatus scores the singing ability by analyzing the input singing voice signal. This scoring tends to be high when the level of the singing voice signal is high and low when it is low. Therefore, even if a person whose singing voice signal is at a low level sings better than a person who is at a high level, the former singing ability may be scored lower than the singing ability of the latter.

As one method for solving such a problem, it is conceivable to use data indicating the state of singing other than the singing voice signal for scoring the singing ability.
As a karaoke apparatus that uses data other than the singing voice signal for scoring, a karaoke apparatus disclosed in Patent Document 1 can be cited. This karaoke apparatus photographs the appearance of a singer with a camera, compares still image data captured from the camera with reference image data stored in advance, and scores the singer's appearance such as choreography.
JP 2000-29483 A

However, the karaoke apparatus disclosed in Patent Literature 1 enables scoring of dancers and movements such as dance, and the singing ability is scored by a conventional method. Therefore, it does not help solve the above problem.
The present invention has been made in view of the above-described circumstances, and an object of the present invention is to provide a karaoke apparatus capable of obtaining a scoring result suitable for the skill of a song regardless of the level of a singing voice signal.

The present invention provides a karaoke apparatus that performs an automatic performance based on music data, a shape detection unit that detects the shape of the singer's mouth from a captured image of a camera that images the singer's mouth, and the music data. There is provided a karaoke apparatus having shape scoring means for specifying appropriateness of a shape detected by the shape detecting means and generating image score data representing the specified appropriateness.
In this karaoke apparatus, the shape of the singer's mouth is detected from the captured image of the camera, the appropriateness of this shape is specified based on the music data, and image score data representing the specified appropriateness is generated.

Further, the present invention provides an aspect ratio measuring means for measuring an aspect ratio of the singer's mouth from a captured image of a camera that images a singer's mouth in a karaoke apparatus that performs automatic performance based on music data, and the music Provided is a karaoke apparatus having an aspect ratio scoring unit that specifies appropriateness of an aspect ratio measured by the aspect ratio measuring unit based on data and generates image score data representing the specified appropriateness.
According to this karaoke apparatus, the aspect ratio of the singer's mouth is measured from the captured image of the camera, the appropriateness of this aspect ratio is specified based on the music data, and image score data representing the specified appropriateness is generated Is done.

Further, the present invention provides a karaoke apparatus that performs an automatic performance based on music data including lyrics data, and a shape detection unit that detects a shape of the singer's mouth from a captured image of a camera that images the singer's mouth; Storage means storing shape data representing an appropriate shape of the mouth for each vowel, vowel specifying means for specifying a vowel to be pronounced by the singer from the lyrics data, and vowels specified by the vowel specifying means A karaoke apparatus comprising: shape scoring means for identifying appropriateness of the shape detected by the shape detecting means based on the shape data stored in the storage means, and generating image score data representing the specified suitability I will provide a.
According to this karaoke apparatus, the vowel that the singer should pronounce is identified from the lyrics data, while the shape of the singer's mouth is detected from the captured image of the camera. Then, the appropriateness of the shape is specified based on the shape data for the specified vowel, and image score data representing the specified appropriateness is generated.

Further, the present invention provides a karaoke apparatus that performs automatic performance based on music data, a size measuring unit that measures the size of the singer's mouth from a captured image of a camera that images the singer's mouth, and the music data A karaoke apparatus having size change scoring means for specifying appropriateness of temporal change in the size measured by the size measuring means and generating image score data representing the specified suitability is provided.
According to this karaoke apparatus, the size of the singer's mouth is measured from the captured image of the camera. Then, the appropriateness of the temporal change in the measured size is specified based on the music data, and image score data representing the specified appropriateness is generated.
According to each of the above karaoke apparatuses, since the image score data is generated without using the singing voice signal, a scoring result corresponding to the skill of the song can be obtained regardless of the level of the singing voice signal.

In addition, a voice scoring means for scoring the singer's singing ability using a signal representing the singing voice of the singer and generating voice score data representing a scoring result for each karaoke device, and the image score data And a total scoring means for generating total score data representing the singing ability of the singer using the voice score data.
In this case, the singer's singing ability is scored using a signal representing the singing voice of the singer, and voice score data representing the scoring result is generated. And the total score data showing a singer's singing power is produced | generated using this audio | voice score data and image score data. Thus, since the total score data is generated using the image score data generated without using the singing voice signal, a scoring result suitable for the skill of the song is obtained regardless of the level of the singing voice signal. Can do. Moreover, since the total score data is generated not only using the image score data but also using the voice score data representing the scoring result using the signal representing the singing voice, the scoring result indicating the singing skill with high accuracy is obtained. be able to.

Embodiments of the present invention will be described below with reference to the drawings. In each figure, the same code | symbol is attached | subjected to the common part.
FIG. 1 is a diagram showing a hardware configuration of a karaoke apparatus according to an embodiment of the present invention.
This karaoke device has a function of performing automatic performance based on music data and a function of scoring a singer's singing ability. The figure collects the singer's singing voice and outputs a singing voice signal. A microphone 1 is also shown.

  This karaoke device includes a ROM (Read Only Memory) 2 in which an initial program is written, a CPU (Central Processing Unit) 3 that reads and executes the program, a RAM (Random Access Memory) 4 that is used as a work area of the CPU 3, and a song A hard disk 5 in which data and system programs are written, a voice input I / F (interface) 6 that inputs a singing voice signal output from the microphone 1 and supplies it to the CPU 3, and an instruction input unit 7 that supplies a performance start instruction to the CPU 3 And a display panel 8 for displaying a scoring result, and a camera 9 for imaging the singer's face and supplying image data representing the captured image to the CPU 3.

FIG. 2 is a diagram schematically showing the composition of music data.
The music data is composed of data such as a header, musical tone data, guide melody data, and lyrics data. The header is bibliographic data related to music such as a music number used for automatic performance. The musical tone data is sequence data including a plurality of event data and duration data indicating the time interval of each event, and automatic performance is performed based on the musical tone data. The guide melody data is melody sequence data to be sung by the singer. In a general karaoke apparatus that scores singing ability, singing ability is scored based on the pitch, volume, rhythm, and tempo represented by the guide melody data. The lyric data is sequence data of lyrics to be sung by the singer. In a general karaoke apparatus, lyrics are displayed using the lyrics data, and the display color of the lyrics is changed according to the progress of the music.

  In FIG. 1, a camera 9 is a camera that automatically adjusts the orientation, zoom, and focus so that the image of the singer's face is large in the center, and imaging and supply of image data to the CPU 3 are determined in advance. At specified time intervals. The CPU 3 executes an initial program when a power supply (not shown) is turned on, and then executes a system program. A performance program for performing an automatic performance based on music data and a scoring program for scoring a singer's singing ability are written on the hard disk 5. The CPU 3 executing the system program stores the instruction input unit 7. When a performance start instruction is input, the performance program and the scoring program are read out and executed.

  Of these programs, the only new one is the scoring program. Therefore, below, the part which concerns on scoring of a singer's singing power is demonstrated among the functional structures and operation | movement of a karaoke apparatus when CPU3 is executing the scoring program. However, since three types of embodiments with different contents of the scoring program are assumed here, these embodiments will be described in order.

[First Embodiment]
FIG. 3 is a diagram showing a functional configuration of the karaoke apparatus 10 according to the first embodiment of the present invention. The karaoke apparatus 10 includes a camera 9, an aspect ratio measuring unit 11, an aspect ratio scoring unit 12, a voice scoring unit 13, a total scoring unit 14, and a score display unit 15.
The aspect ratio measurement unit 11 measures the aspect ratio of the singer's mouth using the image data supplied from the camera 9. The aspect ratio of the mouth is the ratio of the vertical width to the horizontal width of the area surrounded by the lips. The width / length of the area is the maximum length of the area in the width / length direction of the mouth.

The aspect ratio scoring unit 12 specifies the scoring timing with reference to the guide melody data in the music data being automatically played. The aspect ratio scoring unit 12 performs the aspect ratio scoring process described later at the specified scoring timing. In the aspect ratio scoring process, the aspect ratio scoring unit 12 scores the singer's singing ability using the aspect ratio measured by the aspect ratio measuring unit 11, and generates image score data representing the result of the scoring. The scoring timing is a point in time during which the singer should speak.
The voice scoring unit 13 analyzes the input singing voice signal, scores the singer's singing ability, and generates voice score data representing the result of the grading. The music data is referred to at the time of scoring.

The overall scoring unit 14 accumulates the image score data generated by the aspect ratio scoring unit 12 and the audio score data generated by the audio scoring unit 13. The accumulated score data is discarded when the scoring process ends, that is, when the CPU 3 finishes executing the scoring program. The comprehensive scoring unit 14 performs a later-described comprehensive scoring process at a timing at which the scoring result should be displayed. In this total scoring process, the total scoring unit 14 scores the singer's singing ability using the accumulated image score data and voice score data, and generates total score data representing the result of this scoring.
The score display unit 15 causes the display panel 8 to display a scoring result represented by the total score data generated by the total scoring unit 14.

In the karaoke apparatus 10, image data is output from the camera 9 at a predetermined time interval. Therefore, the aspect ratio measuring unit 11 repeats the process of measuring the aspect ratio of the singer's mouth (see FIG. 4).
In this process, the aspect ratio measurement unit 11 first extracts the outline of the singer's mouth from the captured image represented by the supplied image data (step SA1). This contour line is also a contour line of an image (hereinafter referred to as an aperture image) surrounded by a lip image in the captured image. The contour line can be encoded by any method, but in this embodiment, the contour line is encoded by a chain code indicating the direction of the line segment of the unit length. Next, the vertical width and the horizontal width of the aperture image are measured based on the contour line and the orientation of the captured image (step SA2: YES, SA3). The vertical / horizontal width of the aperture image is the maximum length of the aperture image in the vertical / horizontal direction of the captured image. For example, when the relationship between the captured image and the outline is as schematically shown in FIG. 5, the vertical width of the aperture image is H and the horizontal width is W. Next, the ratio of the vertical width to the horizontal width is calculated, and this is regarded as the aspect ratio of the singer's mouth (step SA4). If the singer's mouth is closed, the extraction of the contour line fails, so the aspect ratio measuring unit 11 regards zero as the singer's mouth aspect ratio (step SA2: NO, SA5).

  On the other hand, the aspect ratio scoring unit 12 performs the aspect ratio scoring process of the flow shown in FIG. 6 when the scoring timing comes. In the aspect ratio scoring process, the aspect ratio scoring unit 12 determines whether or not the singer's mouth is open by using the aspect ratio measured by the aspect ratio measuring unit 11 at this time (step SB1). This determination is performed, for example, by comparing the measured aspect ratio with a predetermined threshold value. When the singer is open, the aspect ratio measured by the aspect ratio measuring unit 11 at this time is relatively large, so the aspect ratio scoring unit 12 determines that the singer's mouth is open. Then, image score data representing a predetermined score is generated (step SB2). This score is a positive value. On the contrary, when the singer is closed, the aspect ratio measured by the aspect ratio measuring unit 11 at this time is relatively small, so the aspect ratio scoring unit 12 has the singer's mouth closed. And the image score data representing the zero point is generated (step SB3).

  In parallel with the above processing, the voice scoring unit 13 analyzes the input singing voice signal, scores the singer's singing ability, and generates voice score data representing the result of the grading. The scoring result represented by the voice score data is worse than the singer's skill when the level of the singing voice signal is low. The image score data and audio score data generated in this way are accumulated by the total scoring unit 14.

And when the timing which should display a scoring result comes, the comprehensive scoring part 14 performs the comprehensive scoring process of the flow shown in FIG. In the total scoring process, the total scoring unit 14 adds the score determined from the image score data to the score determined from the accumulated voice score data (step SC1). Next, it is determined whether or not the addition result exceeds a predetermined full score (step SC2), and if it exceeds, total score data representing a full score is generated (step SC3). On the contrary, if not exceeding, total score data representing the addition result is generated (step SC4).
Then, when it is time to display the scoring results, the scoring results represented by the total score data are displayed on the display panel 8 by the score display unit 15.

  As is clear from the above, according to the karaoke apparatus 10, if the singer opens his mouth correctly at the scoring timing, the total score rises within a range not exceeding the full score. If you are good at singing, you can open your mouth at the right time, so in a general karaoke device, the level of singing voice signal is low, so the singing ability is scored lower than the singing skill. In addition, by using the karaoke apparatus 10, it is possible to obtain a scoring result commensurate with the skill of the song.

[Second Embodiment]
FIG. 8 is a diagram showing a functional configuration of the karaoke apparatus 20 according to the second embodiment of the present invention. The karaoke apparatus 20 is different from the karaoke apparatus 10 in FIG. 1 in that it has a size measuring section 21 and a size scoring section 22 instead of the aspect ratio measuring section 11 and aspect ratio scoring section 12, and ideal change data 23 to be described later. Is written in the RAM 4. The ideal change data 23 is written into the RAM 4 when the scoring program is started.

  The ideal change data 23 is data used in a size scoring process, which will be described later, and has envelope data representing an ideal temporal change in mouth size for each case. The reason why the envelope data exists for each case is that the ideal temporal change of the mouth size differs for each case. For example, the ideal temporal change in mouth size differs at the start of utterance and at the end of utterance.

The size measuring unit 21 measures the size of the singer's mouth using the image data supplied from the camera 9 and generates size data representing the measurement result.
The size scoring unit 22 accumulates the size data generated by the size measuring unit 21. Further, the size scoring unit 22 specifies the scoring timing with reference to the guide melody data in the music data being automatically played. In addition, the size scoring unit 22 performs a size scoring process described later at the specified scoring timing. In this size scoring process, the size scoring unit 22 scores the singer's singing ability using the accumulated size data, and generates image score data representing the result of the scoring.

  In the karaoke apparatus 20, image data is output from the camera 9 at a predetermined time interval. Therefore, the size measurement part 21 repeats the process (refer FIG. 9) which measures the magnitude | size of a singer's mouth and produces | generates size data. This process differs greatly from the process shown in FIG. 5 in that it does not measure the aspect ratio, but measures the size (for example, area) of the aperture image and generates size data representing the measurement result (step SD1). SD2). If the contour extraction fails, the size measuring unit 21 generates size data representing zero (step SD3). The size data also includes data indicating the generation time.

  On the other hand, the size scoring unit 22 performs the size scoring process of the flow shown in FIG. 10 when the scoring timing comes. In this size scoring process, the size scoring unit 22 first calculates a change adaptability indicating the appropriateness of the temporal change in the actual mouth size of the singer with respect to the ideal temporal change in the mouth size. (Step SE1). Specifically, first, event data that has just reached the playback timing is extracted from the guide melody data in the music data that is being played automatically, a case is specified based on this event data, and an envelope corresponding to the specified case is extracted. Data is extracted from the ideal change data 23. Next, the change adaptability is calculated by comparing the temporal change represented by the extracted envelope data with the temporal change represented by the accumulated size data. The more similar the two changes are, the higher the degree of change fit is.

  Next, the size scoring unit 22 generates image score data representing the image score according to the calculated change adaptability (step SE2). The generated image score data is accumulated by the total scoring unit 14 and used in the total scoring process. Note that the higher the degree of change suitability, the higher the image score. However, the lowest point of the image score is set to zero, and the image score does not become a negative value no matter how low the change adaptability is.

In parallel with the above processing, voice score data is generated by the voice scoring unit 13, and the generated voice score data is accumulated by the total scoring unit 14.
Then, when it is time to display the scoring result, the comprehensive scoring unit 14 performs the above-described comprehensive scoring process to generate the total score data, and the scoring result represented by this total score data is displayed on the display panel 8. The

  As is clear from the above, according to the karaoke apparatus 20, if the singer sings while changing the size of the mouth correctly, the total score increases within a range not exceeding the perfect score. If the singer is a good singer, singing with the mouth size changed correctly, the singing voice signal level is low in a general karaoke device, so the singing ability is lower than that of a singer. Even a scored person can use the karaoke device 20 to obtain a scoring result commensurate with the skill of the song.

[Third Embodiment]
FIG. 11 is a diagram showing a functional configuration of the karaoke apparatus 30 according to the third embodiment of the present invention. The karaoke apparatus 30 differs from the karaoke apparatus 10 of FIG. 1 in that it has a shape scoring unit 31 instead of the aspect ratio scoring unit 12 and ideal shape data 32 is written in the RAM 4. The ideal shape data 32 is written into the RAM 4 when the scoring program is started. The ideal shape data 32 has shape data representing the ideal aspect ratio of the mouth for each vowel, and is used in the shape scoring process described later.

  The shape scoring unit 31 specifies the scoring timing with reference to the lyrics data in the music data being automatically played. In addition, the shape scoring unit 31 performs a later-described shape scoring process at the specified scoring timing. In this shape scoring process, the shape scoring unit 31 scores the singer's singing ability using the aspect ratio measured by the aspect ratio measuring unit 11, and generates image score data representing the result of the scoring.

In the karaoke apparatus 30, the aspect ratio measurement part 11 repeats the process (refer FIG. 4) which measures the aspect ratio of a singer's mouth.
On the other hand, the shape scoring unit 31 performs the flow shape scoring process shown in FIG. 12 when the scoring timing comes. In this shape scoring process, the shape scoring unit 31 first calculates the shape adaptability, which is the adaptability of the aspect ratio measured by the aspect ratio measuring unit 11 at this time with respect to the ideal aspect ratio. More specifically, the syllable to be pronounced by the singer is specified based on the lyrics data in the music data being automatically played (step SF1), the vowels constituting this syllable are specified (step SF2), The corresponding shape data is extracted from the ideal shape data 32 (step SF3), and the shape conformity is calculated using the ideal aspect ratio represented by the shape data (step SF4).

  Next, the shape scoring unit 31 generates image score data representing the image score according to the calculated shape suitability (step SF5). The image score data generated by the shape scoring unit 31 is accumulated by the comprehensive scoring unit 14 and used in the comprehensive scoring process. Note that the higher the shape matching degree, the higher the number of image points. However, the lowest point of the image score is set to zero, and the image score does not become a negative value no matter how low the shape conformity is.

In parallel with the above processing, voice score data is generated by the voice scoring unit 13, and the generated voice score data is accumulated by the total scoring unit 14.
Then, when it is time to display the scoring result, the comprehensive scoring unit 14 performs the above-described comprehensive scoring process to generate the total score data, and the scoring result represented by this total score data is displayed on the display panel 8. The

  As is clear from the above, according to the karaoke apparatus 30, if the singer sings with the mouth opened in the correct shape at the scoring timing, the total score rises in a range not exceeding the full score. If you are a good singer, you can sing with your mouth open in the correct shape, so in a general karaoke device, the singing voice signal level is low, so the singing ability is lower than that of a singer. Even if it exists, the scoring result commensurate with the skill of the song can be obtained by using the karaoke device 30.

As described above, according to the embodiment of the present invention, even if the singing voice signal level is low, it is possible to obtain a scoring result suitable for the skill of the song, so it is desired to sing with the minimum volume that can be heard around. The singer raises his voice against his will, the singer who wants to sing far away from his mouth and the singer wants to sing his voice against his mouth, and the singer selects the microphone to use with the microphone sensitivity in mind There is no need to do.
In addition, in the embodiment of the present invention, the number of image points is used only for adding in consideration of individual differences in how to move the mouth during singing. In other words, even if the way of singing is not ideal, the score of singing ability will not be unduly lowered.

[Deformation]
In the first embodiment described above, it is determined whether or not the mouth is open by using the aspect ratio of the mouth, but this determination method is arbitrary. For example, the determination may be made using the area of the opening image, or it may be determined that the mouth is open when the contour line has been successfully extracted.

In the first and second embodiments described above, it is assumed that the singer sings with the face facing the camera 9, and the vertical width / width in the captured image is regarded as the vertical width / width of the mouth. When such a premise does not exist, a three-dimensional model of the mouth is generated from the captured image, the direction corresponding to the direction of the mouth in the three-dimensional space is defined as the vertical direction / horizontal direction, and the vertical direction / The maximum length of the three-dimensional opening region in the horizontal direction may be obtained as the vertical width / horizontal width of the mouth. In the second embodiment, a three-dimensional model of the mouth may be generated from the captured image, and the size of the opening area on the surface corresponding to the direction of the mouth in the three-dimensional space may be obtained as the size of the mouth.
In the first and third embodiments described above, the aspect ratio of the mouth is used for comparison, but encoded data of the contour line may be used.

Further, in the third embodiment described above, temporal changes are compared with each other, and image score data having a value corresponding to the change suitability is generated, but the vowel of the sound sung by the singer from the captured image The vowels are compared with the vowels determined from the music data, and if the two coincide, the image score data indicating a positive value is generated, and if the two do not match, the image score data indicating zero is generated. It may be.
In the above-described embodiment, the aspect ratio and size are obtained after obtaining the contour line of the aperture image. However, the upper, lower, left, and right ends of the aperture image are detected, and the positions of these ends in the actual image are used. Thus, the aspect ratio and size may be obtained.

  Moreover, although the camera 9 which images a singer's face was used in embodiment mentioned above, you may use the camera which images only a singer's mouth and the circumference | surroundings of a microphone. In short, it is only necessary to image the singer's mouth. A camera that cannot automatically adjust the orientation, zoom, and focus may be used. In this case, the camera direction and zoom should be set so as to capture the range in which the singer's face or mouth can be located. Further, an image input I / F for inputting image data output from the camera may be provided in the karaoke device, and the camera and the karaoke device may be configured separately.

In the above-described embodiment, the image scoring data is generated using music data having a general configuration. However, the scoring data is embedded in the music data, and the image scoring data is generated using the data. It may be.
In the above-described embodiment, the scoring timing is specified based on the music data. However, a form in which the scoring timing is specified using the captured image is also conceivable. For example, a point in time when the mouth of the singer starts to close or a point in time when the mouth of the singer starts to close may be detected as a scoring timing, and the appropriateness of how to close the mouth may be scored.
In the embodiment described above, the processing related to scoring is performed by the CPU 3, but a DSP (Digital Signal Processor) may be provided so that part or all of the processing is performed by the DSP.

  By the way, since moving the mouth well leads to good utterance, the karaoke apparatus according to the above-described embodiment may be used for practicing how to move the mouth. In this case, in order to tell the singer how to move the mouth, it can be modified to display the result of scoring using the captured image as it is, or negative image score data can be generated It is desirable to deform.

It is a figure which shows the hardware constitutions of the karaoke apparatus which concerns on embodiment of this invention. It is a figure which shows typically the structure of the music data which concern on the same embodiment. It is a figure which shows the function structure of the karaoke apparatus 10 which concerns on 1st Embodiment of this invention. It is a flowchart which shows the flow of the process which the aspect ratio measurement part 11 performs. It is the figure which illustrated typically the relationship between a captured image and an outline. It is a flowchart which shows the flow of an aspect ratio scoring process. It is a flowchart which shows the flow of a comprehensive scoring process. It is a figure which shows the function structure of the karaoke apparatus 20 which concerns on 2nd Embodiment of this invention. It is a flowchart which shows the flow of the process which the size measurement part 21 performs. It is a flowchart which shows the flow of a size scoring process. It is a figure which shows the function structure of the karaoke apparatus 30 which concerns on 3rd Embodiment of this invention. It is a flowchart which shows the flow of a shape scoring process.

Explanation of symbols

DESCRIPTION OF SYMBOLS 1 ... Microphone, 2 ... ROM, 3 ... CPU, 4 ... RAM, 5 ... Hard disk, 6 ... Voice input I / F, 7 ... Instruction input part, 8 ... Display panel, 9 ... Camera, 10, 20, 30 ... Karaoke Device: 11: Aspect ratio measuring unit, 12: Aspect ratio scoring unit, 13: Audio scoring unit, 14: Comprehensive scoring unit, 15 ... Score display unit, 21 ... Size measuring unit, 22 ... Size scoring unit, 31 ... Shape scoring unit Department.

Claims (5)

In a karaoke device that performs automatic performance based on music data,
A shape detecting means for detecting the shape of the singer's mouth from a captured image of a camera that images the singer's mouth;
A karaoke apparatus comprising: shape scoring means for specifying appropriateness of the shape detected by the shape detecting means based on the music data and generating image score data representing the specified suitability. In a karaoke device that performs automatic performance based on music data,
An aspect ratio measuring means for measuring an aspect ratio of the singer's mouth from a captured image of a camera that images the singer's mouth;
A karaoke apparatus comprising: aspect ratio scoring means for specifying appropriateness of the aspect ratio measured by the aspect ratio measuring means based on the music data, and generating image score data representing the specified appropriateness. In a karaoke device that performs automatically based on music data including lyrics data,
A shape detecting means for detecting the shape of the singer's mouth from a captured image of a camera that images the singer's mouth;
Storage means for storing shape data representing an appropriate shape of the mouth for each vowel;
Vowel identification means for identifying a vowel that the singer should pronounce from the lyrics data;
For the vowels specified by the vowel specifying means, the appropriateness of the shape detected by the shape detecting means is specified based on the shape data stored in the storage means, and image score data representing the specified appropriateness is generated. Karaoke apparatus having shape scoring means. In a karaoke device that performs automatic performance based on music data,
Size measuring means for measuring the size of the singer's mouth from a captured image of a camera that images the singer's mouth;
A karaoke apparatus comprising: a size change scoring unit that identifies appropriateness of a temporal change in size measured by the size measuring unit based on the music data, and generates image score data representing the identified suitability. A voice scoring means for scoring the singing ability of the singer using a signal representing the singing voice of the singer, and generating voice score data representing a scoring result;
5. A total scoring unit that generates total score data representing the singing ability of the singer using the image score data and the voice score data. 6. Karaoke equipment.

Images