JP2008146092A - Karaoke device - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To give indication of improvement of singing ability in a part which is required for a singing person. <P>SOLUTION: A karaoke device of the invention which performs automatic playing on the basis of music piece data for expressing a music piece, comprises: a conversion means for converting voice of the singing person to song voice data; a song voice data storing means for storing the song voice data for each period by dividing the music piece into a prescribed period; an evaluation means for evaluating the song voice data by comparing them with reference data for each period; and a notification means which notifies the evaluation by the evaluation means, and which displays length from start to end of the music piece by dividing it at a prescribed part, while displaying the evaluation by the evaluation means for each part. <P>COPYRIGHT: (C)2008,JPO&INPIT

Description

  The present invention relates to a karaoke apparatus that can determine the skill of a singer's singing.

  A karaoke apparatus having a function for evaluating a voice of a singer uttered according to a karaoke performance and displaying a scoring result has been put into practical use. In general, in this type of karaoke apparatus, the pitch and pitch (each note is extracted from guide melody data in the MIDI (Musical Instrument Digital Interface: registered trademark) format, which is included in karaoke music data and serves as a singing standard). Length), pitch, and other elements are compared with each element extracted from the singing voice data obtained by digitizing the voice of the singer, and an evaluation method is used to evaluate the degree of deviation. As a result, the score obtained through the entire music, or the rank or deviation value calculated based on the score (see, for example, Patent Document 1).

JP 2002-221973 A

  However, in this evaluation method, although the skill of the entire song is indicated by the score, it is unknown which part the singer has specifically dropped, and knows which part improves the score It is difficult.

  The present invention has been made under such a background, and an object of the present invention is to provide a karaoke apparatus that gives a suggestion of improving singing ability to a part necessary for a singer.

  The karaoke apparatus according to the present invention is a karaoke apparatus that performs an automatic performance based on music data representing music, conversion means for converting a singer's voice into singing voice data, and dividing the music into predetermined sections, for each section. Singing voice data storage means for storing the singing voice data, evaluation means for evaluating the singing voice data in comparison with reference data for each section, and notification means for notifying the evaluation by the evaluation means. And a notification means for displaying the length from the start to the end of the music by dividing it into predetermined parts and displaying the evaluation by the evaluation means for each of the parts.

  According to the present invention, it is possible to give a suggestion for improving the singing ability to a singer on a necessary part.

  Embodiments of the present invention will be described below with reference to the drawings.

<Configuration of Embodiment>
FIG. 1 is a view showing the appearance of an embodiment of the present invention. In the figure, reference numeral 1 denotes a karaoke apparatus main body. A monitor 2 is connected to the karaoke apparatus main body 1, speakers 3 and 3, and a microphone 4 are connected to the left and right. A remote control terminal 5 is a device for remotely operating the karaoke apparatus body 1 by infrared rays.

  FIG. 2 is a block diagram showing an electrical configuration of the karaoke apparatus main body 1, and each part will be outlined below.

  A CPU (Central Processing Unit) 11 uses a RAM (Random Access Memory) 13 as a work area, and controls various parts of the apparatus by executing various programs stored in a ROM (Read Only Memory) 12. The RAM 13 includes a MIDI buffer for temporarily storing MIDI data.

  A communication I / F (interface) 15 receives music data from the host computer 6 that is a music data distribution source, and transfers the music data to an HDD (Hard Disk Drive) 14 under the control of the CPU 11. Data transfer to the HDD 14 by DMA (Direct Memory Access) is also possible. The song data includes various information related to the song, such as a song number, song name, genre, release date, song performance time, etc., and a performance indicating the performance sound of the song divided into a plurality of tracks. Data and lyrics data indicating lyrics are included. The performance data also includes guide melody data indicating the melody of the music.

  The HDD 14 stores music data, scoring data, and digital data of singing voices input from the microphone 4. The scoring data is data indicating the result of scoring the singing voice of the singer obtained by the scoring program executed by the CPU 11.

  The singing voice of the singer input from the microphone 4 is output from the speakers 3 and 3 via the amplifier 21. Also, the singing voice is A / D converted by the voice processing DSP 20 and used as a singing voice data in the scoring process, and is stored in the HDD 14 as described above.

  The operation unit 16 detects a signal generated by an input operation to the operation panel or the remote control terminal 5 in the karaoke apparatus body 1 and outputs the detection result to the CPU 11.

  The display control unit 17 performs control for causing the monitor 2 to display video data, lyrics subtitles, and scoring results.

  The tone generator 18 synthesizes a musical tone signal corresponding to the music data output from the CPU 11 and outputs the synthesized signal to the effect DSP 19 as a karaoke performance sound.

  The effect DSP 19 gives effects such as reverb and echo to the karaoke performance sound generated by the sound source device 18. The effected karaoke performance sound is D / A converted by the effect DSP 19, synthesized and amplified by the amplifier 21 with the singing voice input from the microphone 4, and output to the speakers 3 and 3.

<Operation of Embodiment>
Next, the operation of the karaoke apparatus having the above configuration will be described.

First, when the user performs a music specifying operation on the operation unit 16, the music data of the specified music is transferred to the MIDI buffer of the RAM 13, and the CPU 11 sequentially reads out the transferred music data. The reproduction of the music is started. CPU11 acquires the performance time of a music from the header of music data at the time of the start of a music, and divides the whole music into a certain number of blocks (it is 30 in this embodiment) (refer FIG. 3a). The CPU 11 manages the start time T _n (n = 1, 2, 3,..., 30) of each block B _n (n = 1, 2, 3,..., 30).

  When the singer starts singing, the singer's singing voice signal is input from the microphone 4 in parallel with the reproduction of the music, and is input to the voice processing DSP 20. The voice processing DSP 20 A / D converts the input singing voice signal, and the A / D converted singing voice data is stored in the HDD 14 in units of blocks.

  On the other hand, based on the guide melody data in the music data sequentially read out from the MIDI buffer of the RAM 13, a signal of a predetermined tone color is generated by the sound source device 18, thereby generating a melody sound, and the melody to be sung during singing is uncertain. If necessary, the singer can listen to this melody and confirm it as necessary. In the present embodiment, the score is obtained by the voice processing DSP 20 performing comparison / evaluation processing on the guide melody data and the singing voice data. The score calculation process in this embodiment is as follows.

  On the basis of the guide melody data and the singing voice data, the voice processing DSP 20 extracts information on the pitch and the pitch, and compares them. The pitch data and the tone length data extracted from the guide melody data are hereinafter referred to as reference data. FIG. 4 shows a flowchart of the comparison process regarding the pitch and the tone length. FIG. 5 illustrates a singer's guide melody and singing voice as a musical score.

  First, it is determined whether both reference data and singing voice data have non-zero values (step S01). In the case of YES, that is, this means that the singer's singing voice data exists in the part that should originally be sung, and in this case, a pitch difference D is obtained (step S02). In FIG. 5, the notes of both the guide melody and the singing voice are shown, but if the guide melody and the singing voice indicate the same scale, the difference is 0. In FIG. 5, a difference occurs in the portion B, and the difference is 0 in the other portions.

In the present embodiment, in order to obtain a pitch difference by a scale of 12 scales instead of a scale of frequency, the difference D is calculated using a unit of Cent (cent), and a minute time Δt (in this embodiment, 50 ms). It is calculated in units) and stored as ΔD. Note that the pitch difference X (cent) of two sounds having frequencies f ₁ and f ₂ is
X = 1200 × log ₂ (f ₁ / f ₂ )
Can be expressed as

  On the other hand, if at least one of the reference data and the singing voice data is not 0 in step S01, that is, if only one of the reference data and the singing voice data is 0 or both are 0, the process proceeds to step S03. In step S03, it is determined which one of these. In the former case, that is, when “sound is not pronounced in the part that is originally pronounced (part A in FIG. 5)” or “sound is pronounced in the part that is not originally pronounced (part C in FIG. 5)”. Yes, these are regarded as abnormal sound lengths, and 1 is recorded in the abnormality ΔE every minute time Δt (step S04). In the latter case, since it is not pronounced in a portion that should not be pronounced, this is normal, and at this time, 0 is recorded in the abnormality ΔE (step S05).

After step S02, step S04, or step S05, the process proceeds to step S06, and a deduction amount at time Δt is calculated. If this deduction amount is ΔP ′,
ΔP ′ = aΔD + bΔE
Is calculated by Here, a and b indicate constants for weighting, and each constant is appropriately set so that the deduction amount becomes an appropriate value. It should be noted that the abnormality ΔE takes only two values of 0 if normal and 1 if abnormal.

The score ΔP of the singer at the minute time Δt is represented by a value obtained by subtracting the above deduction amount ΔP ′ from a perfect score (100 in this embodiment). That is, ΔP = 100−ΔP ′
It becomes.

The score ΔP obtained as described above is averaged for each block, and is sequentially stored and recorded in the HDD 14 as the score P _n of the block. At the same time, the singing voice data of the singer who has been scored is recorded in the HDD 14 in units of blocks in synchronization with the progress of the music. FIG. 6 schematically shows the data inside the HDD 14. The audio data is continuously recorded, but since the time _Tn is recorded, it is possible to easily find the desired block.

When the reproduction of the music is finished, the score P _n of each block is totaled by the CPU 11, and the score of the entire song of the singer is notified to the monitor 2 as shown in FIG. 7, for example, and the evaluation of which part of the song is relative It is also shown whether it was low. In addition, as shown in FIG. 7, a message such as “Let's check by listening to your singing voice!” May be displayed. At this time, by storing a plurality of messages in the HDD 14, it is also possible to selectively display different messages according to the scoring results of the music. For example, if the ratio of the pitch difference ΔD and the abnormality ΔE contributed to the deduction point ΔP ′ is stored, if the ratio of the pitch difference ΔD is high, “the pitch is misaligned between A and B”. It may be displayed more specifically by the user, or when the ratio of abnormal ΔE is high, it may be displayed as “The tempo of singing in parts A and B did not match!” Can give a positive evaluation.

  In accordance with the notification by the monitor 2, the singing voice of the block having a low evaluation is reproduced. Therefore, CPU11 calculates | requires the score for every block of a song, and in order to reproduce | regenerate the song audio | voice of the singer of the part with the low divided | segmented score, it extracts a block. Block extraction is performed as follows.

First, the score of each block is acquired in order to specify the block to be extracted. In this embodiment, in order to extract the part with low evaluation of a song, the block with a small value of _Pn is extracted. In the present embodiment, the threshold value P _t is set, and all the points P _n that are lower than the threshold value P _t are extracted. For example the number P _a in Figure 3a, is P _b corresponds to this. However, if there is no score P _n below the threshold value P _t , the smallest value is extracted from all the scores P _n . The minimum value extracted in this way is hereinafter referred to as a score P _m .

When the score P _m is extracted, the time T _m and time T _{m + 1 of} the corresponding block recorded in the HDD 14, that is, the start time and end time of the block performance are acquired.

When the times T _m and T _{m + 1} are obtained, a reproduction instruction is given to the singing voice data of the singer stored in the HDD 14 corresponding to the block and the karaoke performance sound data corresponding to the block. In the case of reproducing a plurality of blocks, reproduction is performed sequentially from the one with the smaller time T _m , that is, from the beginning of the music. FIG. 3B is a diagram illustrating the reproduction of the two blocks B _a and B _b .

<Modification>
In the present embodiment, the evaluation method is as described above. However, the evaluation method is not particularly limited to the above-described method as long as the score can be calculated in units of blocks, and any method can be used. Good. For example, in the present embodiment, the pitch component and the pitch component of the singing voice are compared with the reference data, and the evaluation is performed by synthesizing both scores. A method of evaluating only one of them as reference data is also conceivable. Further, a simpler evaluation method such as evaluating whether the sound volume is equal to or higher than a predetermined value without using reference data is possible. Further, regarding the definition of the block, in this embodiment, the block is defined by dividing one piece of music into 30 equal parts, but of course, this can be equally divided by other numbers. Various methods can be considered, such as dividing each phrase into measures or dividing into phrases.

  In addition, for the purpose of enhancing entertainment, it is possible to extract a high-scoring block by the same means as in the present embodiment and reproduce it at the time of displaying the score result, or both of them can be reproduced. .

  As a notification method, there are a visual notification method using the monitor 2 and an auditory notification method using the speakers 3 and 3, and they can be used together.

Furthermore, when playing back audio, in order to make it easier for the user to understand the part, instead of playing back only the block as described above, it is possible to play back one or more blocks in succession. Also good. Further, a playback method may be adopted in which the volume is gradually increased from silence at the start (fade in) and gradually decreased (fade out) at the end. In addition, when there are a plurality of playback locations, playback can be performed in the order in which the actual music is played back, or playback can be performed in order from the block with the lower score based on the evaluation result.
Also, instead of setting the playback method on the karaoke device side as described above, for example, a list of blocks to be played back is displayed on the monitor 2, and the user uses the remote control terminal 5 to select a desired block from the list. It is also possible to reproduce an arbitrary portion as many times as possible by giving a reproduction instruction.

  As described above, according to the present invention, it is possible to give suggestions for improving the singing ability with respect to a portion necessary for the singer.

It is a whole block diagram of the karaoke apparatus which is one Embodiment of this invention. It is a block diagram which shows the karaoke apparatus main body in the embodiment. (A) is the figure which illustrated score P (t) with respect to the time t of the singer's song in the embodiment, (b) is the singer reproduced at the time of the announcement of the scoring results after the performance in the embodiment. It is the figure which illustrated singing voice. It is a flowchart for demonstrating the evaluation process of the singing voice data in the embodiment. It is the figure which illustrated a part of guide melody and song voice in the embodiment. It is the figure which illustrated a part of recording content of HDD in the karaoke apparatus. It is the figure which showed an example of the display method of the scoring result in the embodiment.

Explanation of symbols

DESCRIPTION OF SYMBOLS 1 ... Karaoke apparatus main body, 2 ... Monitor, 3 ... Speaker, 4 ... Microphone, 5 ... Remote control terminal, 6 ... Host computer, 11 ... CPU, 12 ... ROM, 13 ... RAM, 14 ... HDD, 15 ... Communication I / F , 16 ... operation unit, 17 ... display control unit, 18 ... sound source device, 19 ... DSP for effect, 20 ... DSP for sound processing, 21 ... amplifier

Claims (1)

In a karaoke device that performs automatic performance based on music data representing music,
A conversion means for converting the voice of the singer into singing voice data;
Singing voice data storage means for dividing the music into predetermined sections and storing the singing voice data for each section;
For the singing voice data, an evaluation means for performing evaluation in comparison with reference data for each section;
Notification means for notifying the evaluation by the evaluation means, wherein the length from the start to the end of the music is divided into predetermined parts and displayed, and the evaluation means for displaying the evaluation by the evaluation means for each part; A karaoke apparatus comprising:

Images