JP2000315089A - Auxiliary voice generating device - Google Patents

Abstract

PROBLEM TO BE SOLVED: To actualize various techniques for cases wherein multiple singers sing artificially by providing the function of a harmonizer, and further generating various auxiliary voices from the singing of one singer and outputting them in various styles. SOLUTION: (1) of a type A is the voice of a main part and a part that a singer actually sings at the time of KARAOKE and (2) is an auxiliary voice generated by the auxiliary voice generating device according to the voice of the main part (1). In a type B, an auxiliary voice (2) is delayed one measure behind the main part (1) similarly to the type A and further converted to a five degrees higher interval. In a type C, an auxiliary voice (2) is outputted one measure after the main part (1), and further respective intervals are increased by five degrees and made twice as long as the original sounds.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an auxiliary voice generating apparatus for generating another auxiliary voice based on a voice actually sung by a singer in a karaoke apparatus or the like.

[0002]

2. Description of the Related Art Some recent karaoke machines have a function called a harmonizer. This means that in addition to the melody line that is the main part sung by the singer, the pitch of the melody line is converted to the pitch of another part such as 3rd, 5th, 3rd, etc., and this is converted to the original voice. This is a function to output in a superimposed manner. With the use of the harmonizer, the person himself can sing the main part and at the same time so-called "harmony" with his own singing voice singing other parts, making karaoke more fun.

[0003]

As described above, the conventional harmonizer generates the sound of the other part from the sound of the melody line, which is the main part, and outputs it in real time, superimposed on the original sound. Was something.
However, techniques for singing together in an actual performance include, as described above, singing by overlapping voices such as 3rd, 5th or 3rd at the same time, as well as vocalization and various other techniques. There are techniques.

The present invention has been made under such a background. In addition to the function of the conventional harmonizer, the present invention further generates various auxiliary sounds from the sound sung by one singer, and generates the auxiliary sounds. It is an object of the present invention to provide an auxiliary sound generation device that can realize various techniques for simulating singing by a plurality of persons by outputting in various modes.

[0005]

In order to achieve the above object, the present invention provides an A / D converter for digitizing an input analog audio signal and an input output from the A / D converter. Input audio storage means for storing audio data, auxiliary audio data generation means for generating auxiliary audio data based on the input audio data, and sequence data for defining in what manner the auxiliary audio data is reproduced. Holding sequence data holding means, auxiliary sound reproduction control means for controlling the reproduction of the auxiliary audio data according to the sequence data, and an analog signal for reproducing the auxiliary audio data reproduced under the control of the auxiliary sound reproduction control means. And a D / A conversion means for converting the sound into an auxiliary sound.

According to a second aspect of the present invention, in the first aspect of the invention, the sequence data holding means holds the sequence data for outputting the input audio data with a predetermined delay.

According to a third aspect of the present invention, in the first aspect of the present invention, the sequence data holding means generates auxiliary audio data in which a pitch of the input audio data is changed by a predetermined frequency, and generates a predetermined timing. It holds the sequence data to be output with only a delay.

According to a fourth aspect of the present invention, in the first aspect of the invention, the sequence data holding means changes the pitch of the input voice data by a predetermined frequency, changes the length of the sound by a predetermined length, Further, it holds sequence data to be output with a delay of a predetermined timing.

According to a fifth aspect of the present invention, in the first aspect of the invention, the sequence data holding means holds sequence data for outputting previously input and stored input audio data at a predetermined timing. It is.

According to a sixth aspect of the present invention, in the first aspect of the invention, the auxiliary voice data generating means divides the voice of the input voice data and performs a predetermined process on each of the divided voices. At the same time, the temporal order of the divided voices is changed, and the sequence data holding means holds the sequence data to be output with a delay of a predetermined timing.

[0011]

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

FIG. 1 is a diagram showing, as musical scores, various auxiliary voices that can be generated by the auxiliary voice generating device of the present embodiment. First, with reference to FIG. 1, what can be realized when the auxiliary sound generation device of the present embodiment is applied to karaoke will be described.

In FIG. 1, type A is a main part of a singer who actually sings at karaoke or the like.
Is a subpart generated by the auxiliary voice generation device based on the voice of the main part. In the A type, the same sound as the main part is played back one bar later than the main part, which is a subpart, which is superimposed on the main part's sound (the sound of the singer singing in real time). Is output. That is, a technique similar to a round singing is realized.

In the B type, as in the A type, the sound of the subpart is reproduced one bar later than the main part, but the sound of the subpart is at a pitch higher by 5 degrees than the sound of the original main part. Is converted and played. Therefore, a special effect different from a simple singing is obtained. In the C type, not only is the sound of the subpart output one bar later than the main part,
The pitch of each sound is changed based on the musical score data, and the length of each sound is made twice as long as the original sound.

In the case of the D type, a hand corresponding to "sole" shown in the above is recorded in advance before the singer actually sings. When the song progresses along with the karaoke as described above and reaches a predetermined location (here, at the end of the second bar), a hand that has been previously recorded and that is “sole” is automatically output. Normally, the hand of the other person is put in by another person,
With this D type, you can get your voice in hand with your own song.

In the E type, the sound of the main part is replaced with the original sound, and the first consonant of each sound is deleted, and only the vowel is output. Therefore, the sound of the subpart is sound irrelevant to most of the main part. This is one example of how to generate auxiliary voices that are not related to the original voice, and the auxiliary voices that are almost irrelevant to the main part are generated as back chorus, for example. It is unique in that it can be used and that it is your voice.

As described above, various auxiliary voices are generated based on the voice of the user singing the main part or the voice previously captured before the singing, and are output at appropriate timing. Various techniques that cannot be realized without singing at the same time can be realized by one voice in a pseudo manner, and other effect voices can be generated, thereby providing a new way of enjoying karaoke that has never been seen before.

Next, various techniques shown in FIG. 1 and methods for realizing sound effects will be described. FIG. 2 is an overall block diagram of the auxiliary voice generation device of the present embodiment for realizing the various techniques shown in FIG.

In FIG. 2, a microphone 10 is a normal microphone that a singer sings with. An audio signal from the microphone 10 is amplified by an amplifier 11, passes through a low-pass filter 12, and is converted into digital data by an A / D converter 13. When the A / D converter 13 converts the audio signal into digital data, the A / D converter 13 outputs an interrupt signal indicating that the digital data has been generated to the CPU constituting the main part of the sequencer 14. When the CPU receives the interrupt signal, it executes an algorithm shown in FIG.
The audio data is transferred to an input audio buffer in the AM 16b, and audio data is temporarily stored therein. This audio data is transferred to a dual port RAM (DPRAM) 15 under the control of the sequencer 14 when the audio buffer becomes full.

The program ROM 16a stores musical score data.
(Also referred to as “sequence data”). Musical score data is data such as a design document for generating music for each song, and information on pitch, sounding timing, volume, which tone to use, reverb, vibrato, echo, etc. The fourteen CPUs have been processed into an easy-to-handle form.

One of the features of the auxiliary voice generating apparatus according to the present embodiment is that, for all songs that can be sung by a singer,
The score data of the main part and the score data of the selectable subparts are stored in the program ROM as one unit.
16a. However, inside the apparatus, the musical score data of these subparts is also treated as being equivalent to the musical score data of the main part and the musical score data of the musical instrument performance for the accompaniment required.
In other words, the main part (singer's singing), accompaniment, and subpart (generated singing) included in one song are each treated like another song. These musical score data need not always be stored in a memory in the apparatus such as the ROM 16a. For example, a karaoke store or the like can use a communication line or the like from a specialized trader storing a large amount of musical score data. It is also possible to send the information individually and provide it.

When the singer selects a specific song, the score data of the song is stored in the sequencer 1 from the program ROM 16a.
4 Further, when any of the techniques shown in FIG. 1 is selected, the musical score data of the subpart corresponding to the technique is also sent from the program ROM 16a to the sequencer 14. When the performance process is started, the sequencer 14 interprets these musical score data and sends to the PCM sound source 17 instruction data in which the contents of the musical score data are interpreted. The PCM sound source 17 reads necessary tone data of an instrument or the like from a ROM 18 in which tone data is stored based on an instruction from the sequencer 14, and outputs a DPRAM.
The singing person's voice data stored in 15 is also read out. The tone color data is, for example, a digitized waveform of a violin sound in the case of a violin.

A RAM for storing audio data
5 is a dual port RA with two input / output ports
M is used because the RAM is written with data from the sequencer 14 and the data is read out by the PCM tone generator 17 so that they may be accessed simultaneously. However, it is not essential to use a dual port RAM for this part, and a normal RAM can be used. This will be described later.

The PCM sound source 17 generates digital data of each specific music by performing an effect process such as a change of a pitch or a volume, a reverb or an echo according to an instruction from the sequencer 14. The data thus obtained is D
A / A converter 19 converts the signal into an analog signal.
After passing through the low-pass filter 20, the signal is amplified by the amplifier 21 and output from the speaker 22.

The display circuit 25 generates a signal for displaying the lyrics of the selected song in accordance with the progress of the song or for displaying a cursor indicating the location of the lyrics to be currently sung to the singer. This signal is sent to a display device 26 such as a CRT, and is displayed together with a background screen selected according to the music. In addition, it is also used for selecting a song on the screen or displaying the status of the apparatus.

Although not shown in FIG. 2, the microphone 10
It is also possible to process the audio picked up by the above as an analog audio through a different path from the above-mentioned path, mix it with the analog signal generated by the D / A converter 19, and output the mixed signal from the speaker 22. The voice output in this manner is a voice obtained by superimposing the voice sung by the singer and the auxiliary voice generated by the circuit of FIG. The CPU constituting the sequencer 14 shown in FIG. 2 also controls the entire apparatus shown in FIG. 2, and controls the execution of the processing algorithms shown in FIG.

FIG. 3 is a main flowchart executed when a singer selects a certain song and sings along with the accompaniment. Here, it is assumed that lyrics appear on the screen of the display device 26 and the singer sings while watching the lyrics.
In FIG. 3, in the initial setting (Step 10), first, necessary settings such as selection of a song to be sung and selection of a technique, and recording of a need to make a match like D type in FIG. Do.

When the initialization is completed, a subroutine of "score display processing" is entered (Step 11). This is a process of displaying music scores and lyrics on the screen in order on the display device 26. More specifically, the process is to display a portion of the current lyrics as the song progresses, or to highlight a portion to be currently sung on the displayed lyrics.

Subroutine of "performance processing" (Step 12)
Is to digitize the voice of the singer input from the microphone 10, store the digital voice data in a memory, reproduce the voice in real time, and further, based on the voice data stored in the memory, an auxiliary part. For example, a process of reproducing the voice of the user is performed. Accompaniment processing is also performed in this performance processing subroutine.

By repeating the sub-routines of "score display processing" and "performance processing" until the end of the music, the performance of one music is completed. Hereinafter, each subroutine of the "score display process" and the "performance process" in FIG. 3 will be described.

FIG. 4 is a flowchart showing the algorithm of the "score display process" subroutine. this is,
It assists the singer so that the singer can sing at karaoke while watching the lyrics. When the lyrics of one screen of a certain song are displayed on the display device 26, the head of the lyrics is first highlighted. Then, the display cursor is moved in accordance with the progress of the song (Step 20), and it is determined whether the cursor has reached the end (Step 21). If the cursor has not reached the end, the display cursor is moved to the next part and updated (Step 22). If the cursor has reached the end, the lyrics for the next one screen are displayed.

FIG. 5 is a flowchart showing the algorithm of the "performance process" subroutine. When the algorithm shown in the figure is started, first, a timer process is performed.
(Step 30). This timer process is a process for determining the tempo of a note. However, the note here is not a note described in a normal score, but a note of a sequencer. In this apparatus, 1/120 second is set as the note tempo, and the routine from Step 31 to Step 37 shown in FIG. 5 is executed once within this time.

Next, it is determined whether or not there is an unprocessed music piece to be processed at the present time (Step 31). As described above, in the apparatus, the main part, accompaniment, and sub-part (each of which is also referred to as each part) in one song are considered to be different songs. This means that it is determined whether there is any unprocessed song in each of these parts of the song. The algorithm shown in FIG. 5 is executed for each of a plurality of parts in one song.

If there is any unprocessed data in Step 31, the musical score data is read (Step 32), and the musical score data is interpreted (Step 33). Then, it is determined whether or not the end of the score has been reached (Step 34).
It is determined whether or not it is the next processing timing (Step 36). Then, when the processing timing comes, the address of the tone color data is set in the work RAM 16b.
(Step 37) and the note data is stored in the work RAM 16b.
(Step 38). Although the address of the timbre data is set in the work RAM 16b in Step 37, the address of the DPRAM 15 in which the voice of the singer inputted from the microphone is stored is actually the address of Step 54 in FIG.
Is written to the address information table in the transfer process of Then, the sequencer 14, which has received the musical score data for controlling the input voice, reads out the address information from the address information table and
Set to.

If the end of the musical score has been reached in Step 34 (if Yes), performance termination processing is performed (Step 35).
Specifically, an end parameter is set in the work RAM 16b.

In Step 31, if there is no more to be processed at the present time, that is, if the processing to be performed at this time has been completed for all songs (if No), based on the value set in the work RAM, Changes in volume and pitch are calculated (Step 39), and the parameters thus obtained are set in the PCM sound source 17 (Step 40), and a series of performance processing subroutines is terminated.

FIG. 6 is a flowchart of the voice input interrupt processing. In this process, the input audio data is
This is a process required when transferring to the audio buffer in the PU, and is executed as an interrupt process of the CPU constituting the sequencer 14. First, when an audio signal is input, A
Data is read from the / D converter into the input audio buffer in the work RAM 16b (Step 50). Next, it is checked whether or not the buffer capacity is full of data (Step 51). If the buffer is not full, this processing ends and the routine exits. On the other hand, if the buffer capacity is full, data smoothing processing is performed (Step 52). This is because if you change the pitch of the voice and output it as it is,
This is a process for eliminating noises such as pops and smoothing the sound.

Next, data is transferred from the input audio buffer in the work RAM 16b to the DPRAM 15 (Step 53), and information of the transferred data is set (St 53).
ep54). The “setting of information of transferred data” specifically refers to address information indicating where to start arranging data in the DPRAM 15 and the work RAM 16b in FIG.
Is to memorize it.

In FIG. 2, the DPRAM 1 is used as a RAM for storing audio data for the above-described reason.
5 was used. However, in general, DPRAMs have a normal RA
It has less storage capacity than M and is more expensive than ordinary RAM. Therefore, it is conceivable to use a normal RAM instead of the DPRAM as a RAM for storing audio data. FIG. 7 is a block diagram showing a configuration of a memory circuit for storing audio data using two ordinary RAMs.

In FIG. 7, since the CPU 14 is a CPU constituting the aforementioned sequencer, it is represented by the same reference numeral 14. The same applies to the PCM sound source 17. The selectors (SL) 40 and 41 selectively output one of the two input address data in response to a control signal from the CPU 14 and transmit the data to the RAM 45 or 46. That is, when audio data is written from the CPU 14 to the RAM 45 or 46, address data for data writing given from the CPU 14 is output,
When the PCM sound source 17 reads out audio data,
The address data for data reading provided from the M sound source 17 is output.

Bus transceivers (BT) 42a, 42
b, 43a, and 43b are used to select whether to pass the input data or not according to a control signal from the CPU 14. 42a and 42b are for the RAM 45, and 43a and 43b are for the RAM 46. Bus transceiver 42a for RAM 45 and bus transceiver 4 for RAM 46
3a sends the audio data provided from the CPU 14 to each RAM. On the other hand, a bus transceiver 42b for the RAM 45 and a bus transceiver 43b for the RAM 46
Is the PCM sound source 1
Send to 7. The inverters 48 and 49 are inserted so that control signals to the two sets of bus transceivers 42a and 42b and 43a and 43b are inverted from each other.

Next, pitch conversion (pitch shift) will be described. Although several types of pitch shift methods have already been proposed, in this embodiment, the CPU
14 adopts a method of performing a pitch shift by software. However, in the present invention, not only the pitch shift is performed in software by the CPU but also other methods can be used. FIG. 8 shows an example thereof.
It is a block diagram in the case of using the pitch shifter made into a chip exclusively.

In FIG. 8, the pitch shifter 50 has a CP
It is arranged between U14 and DPRAM15. The pitch shift 50 is a dedicated chip IC, and the CPU 1
The information on how much the pitch is to be shifted is received from No. 4, the pitch-converted data is generated in accordance with the information, and the data is transferred to the DPRAM 15. In addition, using a general-purpose DSP (Digital Signal Processor),
A pitch shifter may be configured separately from U14.

Although the embodiment of the present invention has been described above, the present invention is not limited to the above embodiment, and various modifications can be made within the scope of the invention, and these are also described in the present invention. Included in the technical scope.

[0045]

As described above, according to the present invention, digitized voice data is stored, and the auxiliary voice data generating means generates various auxiliary voice data based on the voice data, and converts this into analog signal. In addition to the method of generating the auxiliary voice by the auxiliary voice data generating means, the generated auxiliary voice is simultaneously superimposed on the original voice and sung, and the singing is performed by one's own voice. , Back chorus, and various other techniques can be realized in a pseudo manner, thereby providing a new way of enjoying that has not been provided in conventional karaoke apparatuses.

[Brief description of the drawings]

FIG. 1 is a diagram showing, as a musical score, various auxiliary sounds that can be generated by an auxiliary sound generation device according to an embodiment of the present invention.

FIG. 2 is an overall block diagram of an auxiliary voice generation device according to an embodiment of the present invention.

FIG. 3 is a main flowchart executed when a singer selects a song and sings.

FIG. 4 is a flowchart showing an algorithm of a “score display process” subroutine.

FIG. 5 is a flowchart showing an algorithm of a “performance process” subroutine.

FIG. 6 is a flowchart of a voice input interrupt process.

FIG. 7 is a block diagram showing a configuration of a memory circuit for storing audio data using two ordinary RAMs.

FIG. 8 is a block diagram of a circuit for performing a pitch shift using a pitch shifter chipped exclusively.

[Explanation of symbols]

Reference Signs List 10 microphone 11, 21 amplifier 12, 20, low-pass filter (LPF) 13 A / D converter 14 sequencer (CPU) 15 dual-port RAM (DPRAM) 16a program ROM 16b work RAM 17 PCM sound source 18 tone data ROM 19 D / A converter 22 Speaker 25 Display circuit 26 Display device 40, 41 Selector (SL) 42a, 42b, 43a, 43b Bus transceiver (BT) 45, 46 RAM 48, 49 Inverter 50 Pitch shifter

 ──────────────────────────────────────────────────続 き Continued on the front page F-term (reference) XX20 9A001 BB01 BB03 BB04 BB05 DD13 EE02 EE05 HH15 HH16 HH17 HH18 KK43

Claims (6)

[Claims] 1. A / D conversion means for digitizing an input analog audio signal, input voice storage means for storing input audio data output from the A / D conversion means, and Auxiliary audio data generating means for generating auxiliary audio data, sequence data holding means for holding sequence data that defines in what manner the auxiliary audio data is to be reproduced, and storing the auxiliary audio data in accordance with the sequence data. Auxiliary audio reproduction control means for controlling reproduction, and D / A conversion means for converting auxiliary audio data reproduced under the control of the auxiliary audio reproduction control means into an analog signal. Auxiliary voice generator. 2. The auxiliary sound generation device according to claim 1, wherein said sequence data holding means holds sequence data to be output after delaying said input sound data by a predetermined timing. 3. The sequence data holding means generates auxiliary sound data in which a pitch of the input sound data is changed by a predetermined frequency, and holds sequence data to be output with a delay of a predetermined timing. Item 10. The auxiliary sound generation device according to Item 1. 4. The sequence data holding means changes the pitch of the input voice data by a predetermined frequency, changes the length of the sound by a predetermined length, and holds the sequence data to be output with a delay of a predetermined timing. The auxiliary voice generation device according to claim 1, wherein 5. The auxiliary sound generation apparatus according to claim 1, wherein said sequence data holding means holds sequence data for outputting previously input and stored input audio data at a predetermined timing. 6. The auxiliary audio data generating means divides the audio of the input audio data, performs a predetermined process on each of the divided audios, and changes the temporal order of the divided audios. Holding means,
2. The auxiliary voice generation device according to claim 1, wherein the sequence data is output after being delayed by a predetermined timing.