JPH10301567A - Voice control device for electronic musical instruments - Google Patents

Abstract

(57) [Problem] To provide a musical tone control command based on the identified keyword by identifying a keyword of a musical tone control command which is standardized and inputted through a microphone during a performance. It is an object to provide a voice control device for an electronic musical instrument that can be used. SOLUTION: The present invention relates to an electronic musical instrument for controlling a musical tone by operating a manipulator, wherein an input means for voice-inputting a command for musical tone control, and a musical tone control command from a voice pattern input from the input means. A voice recognition means for pattern-recognizing the indicated keyword and converting it to a control signal for tone control corresponding to the keyword and transmitting the control signal to the sound source CPU, and a voice pattern corresponding to each command are stored. A voice recognition dictionary that is referred to when the means pattern-recognizes words that are voice control keywords;
A conversion table referred to when the command recognized by the voice recognition means is converted into a control signal recognizable by the sound source CPU, wherein the sound source CPU controls the musical tone based on the control signal sent from the voice recognition means. It is configured to

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a control device for an electronic musical instrument having a built-in automatic accompaniment function. The present invention relates to a voice control device for an electronic musical instrument that can perform musical tone control in accordance with a recognized voice pattern.

[0002]

2. Description of the Related Art In a conventional electronic musical instrument having a built-in rhythm, in order to control a rhythm such as a fill-in or an intro during a performance, it is necessary to operate a switch or a foot pedal on a control panel. However, there is a problem that the performance is temporarily stopped and the performance becomes unnatural.

When a large number of tones and rhythms are built in, it is difficult to know where the switches for the desired tone and rhythm are located, and the operation timing is deviated.
Because of the small size of the switch, there were times when he made an operation mistake.

Also, when performing an automatic accompaniment with an electronic musical instrument having an automatic accompaniment function, there is a problem that the operation cannot be performed without knowing the chord form in order to perform chord control. was difficult.

On the other hand, recent advances in speech recognition technology have been remarkable. For example, the name of a destination station can be displayed on a display if the name of a destination station is changed, or the destination name can be expressed by a car navigation system. A voice control system has been realized, such as displaying a map of the surrounding area including a karaoke on a display or selecting a karaoke song by voice.

According to the speech recognition technology which has been remarkably advanced in recent years, a standardized short word or a combination of words can be accurately pattern-recognized for about several hundred words if it is used. Has become.

[0007]

SUMMARY OF THE INVENTION The present invention has been made in view of the above circumstances, and identifies a tone control command which is standardized and is inputted through a microphone during a performance, and based on the identified command. It is an object of the present invention to provide a voice control device for an electronic musical instrument that can perform musical tone control by using the electronic musical instrument.

[0008]

According to the present invention, there is provided an electronic musical instrument for controlling a musical tone by operating a manipulator, comprising: an input unit for inputting a command for controlling a musical tone by voice; A voice recognition unit 30 that recognizes a keyword indicating a musical tone control command from a voice pattern, converts the keyword into a control signal for musical tone control corresponding to the keyword, and sends the control signal to the sound source CPU 10; The voice recognition dictionary 30 is matched when the voice recognition means 30 performs pattern recognition of a word serving as a keyword for voice control, and the command recognized by the voice recognition means 30 is recognized by the sound source CPU 10. And a conversion table 33 which is referred to when converting into a control signal which can be transmitted. Configured to be able to control the musical tone based on the control signal.

The voice recognition means 30 and the sound source CPU 10
In this case, when the voice control switching means 3A is turned on, the tone control by voice recognition becomes possible in addition to the control mode by the operator.

The voice recognition dictionary 34 and the conversion table 33 have data of one or more control elements of timbre control, rhythm control, chord control, volume control, and tempo control. It is configured such that any one or more of the sound volume and the tempo can be controlled.

[0011]

According to the present invention, the speech recognition technology is applied to an electronic musical instrument,
The sound control of the electronic musical instrument during the performance is voice-controlled, thereby enabling a natural and seamless performance. That is, changes in tone, rhythm, volume, tempo, etc.
Speaking of "tempo up", the tempo is faster.

Also, for example, when an instruction is given by a voice of a combination of fixed short words such as "volume 1", the volume is set to 1 and voices are used without using limbs during the performance. It is possible to control the musical tone.

For this reason, in the electronic musical instrument of the present invention, the voice recognition means 30 performs pattern recognition of the voice pattern of the musical tone control command input via the input means 13 as a specific command by collating the voice recognition dictionary 34. Then, the corresponding command is converted into a control signal recognizable by the sound source CPU 10 and sent to the sound source CPU 10.

Command setting section 21 of sound source CPU 10
Sets a flag corresponding to the command based on the control signal sent from the voice recognition means 30, or performs processing such as writing to an event buffer, and waits for control by the control unit 22.

At a predetermined timing, the control unit 22 reads out predetermined control data by referring to the musical tone control table 24 based on the data stored in the flag and the event buffer, and sends the control data to the musical tone signal generating unit 5. The tone signal generation unit 5 performs tone control similar to that of a conventional electronic musical instrument to generate a tone.

With this arrangement, it is not necessary for the performer to use the hands and feet to operate the operators when performing the musical tone control during the performance, and the electronic musical instrument is easy to use. In addition, since the control is performed by voice, the performance is not interrupted by the musical tone control, and a natural performance that follows the flow can be performed.

Further, according to the present invention, even if the position of the code form or the operator is not known, for example, the "A minor"
Then, the automatic accompaniment can be performed by voice, such as playing the "Am" chord, so that the electronic musical instrument is easy for beginners to use.

Further, by providing a voice control switching means 3A for controlling on / off of the voice recognition means 30, when the voice control switching means 3A is turned on, a voice control mode is added in addition to the control mode by the conventional operator. When it is turned off, the mode returns to the normal operation mode. As a result, the control mode can be selected according to the performance phase,
It becomes an electronic instrument that is easier to use.

[0019]

FIG. 1 is a schematic block diagram showing the overall configuration of an electronic musical instrument having a voice control device according to the present invention. Hereinafter, an overall configuration of an electronic musical instrument with a voice control device of the present invention will be described with reference to the drawings.

In the figure, a sound source CPU 10 is a ROM
11 to control the entire electronic musical instrument by executing a control program stored in a program memory unit (not shown).
The control parameters are read from a tone control table 24 provided on the tone generator 11 and sent to the tone signal generator 5 to control the tone generation.

That is, the sound source CPU 10 converts parameters such as a key touch obtained by a touch sensor and a timbre obtained by a panel scan into a sounding start command, a sounding ending command, a timbre changing command, and the like. The tone signal is given to the tone signal generator 5 via the controller to control the sound generation. For this reason, a command setting unit 21 and a control unit 22 are implemented in the sound source CPU 10 as software functions.

The command setting section 21 sends an event signal of the operation panel 3 detected by the panel scan circuit 4 and a signal from the voice recognition section 30 such as setting of a flag in accordance with a set tone generation condition and writing to an event buffer. The setting for musical tone control is performed in accordance with the command instructed based on the specified control signal.

At a predetermined timing, the control unit 22 issues a command set by the command setting unit 21,
Based on a flag set in a predetermined area of M12 and data stored in an event buffer, control data and parameters for tone control are read from the tone control table 24 and developed in the tone signal generator 5. It is.

The ROM 11 stores a program memory unit storing a program for operating the sound source CPU 10, a frequency number which is data for forming a musical tone component signal,
It has various control tables and the like in which waveform numbers, envelope waveform numbers, mode data and the like are stored, and also stores various fixed data.

The musical tone control table 24 stores control data and parameters for generating a tone based on a tone number or a chord name selected by operation of an operator or voice recognition, and a pressed note number. , For example, by type of musical instrument such as electric guitar, piano, organ, etc.
Control values and parameters corresponding to the note numbers are stored.

The RAM 12 has the above-described sound source CPU 1
In addition to the temporary storage of a program for operating 0, a work area of the sound source CPU 10, various registers, counters, flags, buffers, and the like for controlling the electronic musical instrument are defined. Further, it has a data area in which necessary data among the data stored in the ROM 11 is transferred and temporarily stored, and a work area for work.

A plurality of registers in which data necessary for sound emission are set in accordance with the setting states of the keys and switches of the operation panel 3 and the command recognized by the voice, and each tone generation of the tone signal generator 5 The RAM 12 is also provided with an assigner memory for storing data for allocating circuits to unused channels, a storage area for temporarily storing musical sound information, and the like.

The voice recognition unit 30 checks a voice pattern input from the microphone 13 and digitally converted by the A / D converter 14 with a voice pattern corresponding to each command stored in the voice recognition dictionary 34. Then, a command indicated by the input voice is specified, a control signal indicating the specified command is generated and sent to the sound source CPU 10.

The voice recognition unit 30 is composed of a voice recognition CPU 31 and a digital signal processor 32 (hereinafter referred to as DSP).
3. Work RAM 35, voice recognition dictionary 34 and A / D
The converter 14 is connected.

The voice recognition CPU 31 controls the DSP 32, sends a predetermined program to the DSP 32, writes and reads data to and from the work RAM 35, and reads pattern data and a voice recognition dictionary stored in the work RAM 35. It controls the collation of each command stored in 34 with the voice pattern, the conversion of the determined command into a control signal, the transmission of the converted signal to the sound source CPU 10, and the like.

The DSP 32 processes the voice input from the microphone 13 and digitally converted by the A / D converter 14 to perform voice recognition, and generates a command signal to be sent to the sound source CPU 10 based on the recognized command. This is transmitted to the sound source CPU 10. That is, C for voice recognition
The voice pattern input and controlled by the PU 31 is written to and read from the work RAM 35.

Further, the pattern data stored in the work RAM 35 is compared with the voice pattern corresponding to each command stored in the voice recognition dictionary 34, and a voice pattern determined as a keyword indicating a specific command is determined. The sound source CPU 10 converts the signal into a signal recognizable as a signal indicating a command, and converts the converted signal into a signal for the sound source CPU.
Send to 10.

The conversion table 33 is capable of recognizing the identification number or command name of the voice pattern recognized as a specific command by the DSP 32 as a signal indicating the command by the sound source CPU 10 in the same manner as a signal obtained as a result of panel scan. 4 is a table that stores data (address) referred to when converting to a proper signal for all controllable commands.

The conversion table 33 is stored in the voice recognition CPU 3.
Under the control of 1, the DSP 32 refers to the command using the command number or command name determined as a result of the voice recognition as a key, and the read data is sent to the CPU 10 for the sound source. The configuration of the conversion table 33 will be described in detail with reference to FIG.

The work RAM 35 temporarily stores a voice pattern inputted through the microphone 13 and converted into a digital signal by the A / D converter 14.
At a predetermined timing, the voice pattern stored in the work RAM 35 is read out by the DSP 32 and collated with the voice pattern of the command stored in the voice recognition dictionary 34.

The voice recognition dictionary 34 stores a fixed keyword for indicating a command, for example, a voice pattern of "fill-in" or "tempo-up", and a unique identification number indicating each command. For example, a voice pattern and a command identification number are stored in association with each other, such as 36 for tempo up and 23 for fill-in.

Accordingly, the detected keyword is sent to the conversion table 33 as an identification number indicating the command. The correspondence table between the voice patterns and the commands stored in the voice recognition dictionary 34 is provided so as to have a one-to-one correspondence with each set value of each operation element provided on the operation panel 3. ing.

That is, in the case of a volume, an audio pattern is prepared so that the volume scale is set to 1 when "Volume 1" is set, and the volume scale is set to 5 when "Volume 5" is set. ing.

The types of commands provided in the voice recognition dictionary 34 and the conversion table 33, that is, the control elements to be controlled by the musical tone include only types of operators such as tone, rhythm, chord, tempo, and volume. However, the elements that can be recognized by voice may be configured to be selectively implemented from each of these elements as necessary.

The microphone 13 is for inputting a control command by voice during the performance, and the A / D converter 14 is for converting the input voice to a digital signal. The digitally converted voice pattern is temporarily stored in the work RAM 35 under the control of the voice recognition CPU 31. The microphone 13 and the A / D converter 14 are provided, for example, inside a keyboard.

The keyboard 1 is used for designating a musical tone to be generated, and is composed of a plurality of keys and a key switch which opens and closes in conjunction with the key press / release operations of these keys. Key press / key release operations of the performer are detected by a touch sensor, and a signal detected under the control of the sound source CPU 10 is supplied to the musical tone signal generator 5.

The keyboard 1 includes a key press detection unit 2 for detecting the pressed state of each key. Performance information generated by key press or key release of the keyboard 1 is temporarily stored in the RAM 1.
CPU 2 for a sound source stored in a predetermined area of
0 and is supplied to a sound channel assignment unit (not shown).

In this embodiment, the case where the keyboard 1 is used as an input device is described as an example. However, the input device is not limited to this, and for example, an instrument other than a keyboard instrument such as a guitar or MIDI An apparatus or the like may be used.

The key press detecting section 2 detects a key press / key release operation of a player, that is, on / off of a key, and transmits the detected on / off information to the musical tone signal generating section 5 together with the key number. is there. The command setting unit 21 temporarily stores the key on / off information in the RAM 12 and reads out the information as needed.

The operation panel 3 is for instructing a change of tone generation conditions such as a tone color and a volume. In addition to a power switch, various switches such as a tone color selection switch, a mode designation switch, a melody selection switch and a rhythm selection switch are provided. Also, a display unit for displaying various information is provided.

The set / reset state of each switch of the operation panel 3 is detected by a panel scan circuit 4 included therein. Data on the set state of the switch detected by the panel scan circuit 4 is a command. It is stored in a predetermined area on the RAM 12 under the control of the setting unit 21.

The operation panel 3 is provided with a voice control switch 3A for switching the voice control, and a display unit for displaying a setting state of each operation element, a setting mode for editing, a state of automatic performance, and the like. .

The voice control switch 3A is a changeover switch for enabling a tone control by voice during a performance in addition to a control mode of a tone by an operator. When the voice control switch 3A is pressed, the voice control switch 3A is activated. Musical tone control becomes possible.

The tone signal generator 5 reads out tone waveform data and envelope data corresponding to the signal sent from the CPU 10 from the waveform memory 6, adds an envelope to the read tone waveform data, and outputs it as a tone signal. Things.

The tone signal output from the tone signal generator 5 is supplied to an amplifier 8 after being converted into an analog signal by a D / A converter 7. For this purpose, a waveform memory 6 for storing waveform data and envelope data is connected to the tone signal generator 5.

The waveform memory 6 is composed of, for example, the ROM 11, and stores waveform data of musical tones and various envelope data corresponding to musical tone components. Envelope data that is selected using the contents of the tone component selection register as an address is stored.

The D / A converter 7 converts an input digital tone signal into an analog tone signal.
The tone signal converted into an analog signal by the A converter 7 is supplied to the amplifier 8.

The amplifier 8 amplifies the analog tone signal supplied from the D / A converter 7 with a predetermined gain. The output of the amplifier 8 is supplied to a speaker 9.

The speaker 9 converts an analog tone signal as an electric signal sent from the amplifier 8 into an acoustic signal. That is, a tone is emitted according to the generated tone signal.

FIG. 2 is a functional block diagram illustrating the configuration and operation of the voice control device of the present invention. Hereinafter, the overall operation of the voice control device of the present invention will be described with reference to the drawings.

In the configuration shown in FIG.
The analog audio signal input from 3 is digitally converted by the A / D converter 14 and sent to the DSP 32.
The DSP 32 is controlled by the voice recognition CPU 31 to temporarily store the input voice pattern in the work RAM 35.

Next, at a predetermined timing, a signal from the sound source CPU 10 is received and the voice recognition CPU 31
Controls the DSP 32 to read out the voice pattern stored in the work RAM 35 and instruct the DSP 32 to perform voice recognition.

In response to this instruction, the DSP 32 checks the read voice pattern against the voice pattern of each command stored in the voice recognition dictionary 34, and determines whether any of the keywords input by voice corresponds to any command. Is determined.

When the type of the command indicated by the voice pattern read from the work RAM 35 is determined,
The DSP 32 converts the signal into a signal recognizable by the sound source CPU 10 as a control signal indicating a command.
And reads the corresponding signal data from the
Send to U10.

Upon receiving the control signal sent from the voice recognition unit 30, the sound source CPU 10 sends it to the command setting unit 2.
Send to 1. The command setting section 21 sets a flag for tone control according to the command based on the transmitted signal.
Alternatively, data is written to the event buffer.

Next, at a predetermined timing, the control section 22 reads a flag for musical tone control and data in an event buffer, reads control data corresponding to the read data from the musical tone control table 23, and To deploy.

As a result, the tone signal generator 5 reads out a predetermined waveform from the waveform memory 6 and emits the tone in accordance with tone generation conditions developed in the tone signal generator.

FIG. 3 is a diagram for explaining the function of the conversion table 33. Hereinafter, the conversion table 33 will be described with reference to the drawings.
The function of will be described.

The item No in FIG. 3A is a diagram showing an example of the identification number when the command pattern-recognized by the DSP 32 is read out from the voice recognition dictionary 34. DSP as shown
If the command recognized by the 32 is, for example, a tone control organ, the identification number is read from the voice recognition dictionary, and the identification number is read from the voice recognition dictionary.

The conversion table 33 converts the identification number (command) read from the voice recognition dictionary 34 into a signal recognizable by the sound source CPU 10. For example, in the case of an organ, the DSP 32 first controls 00 indicating the element (tone color) is read out, and subsequently 02 indicating the control item organ of the tone color is read out and sent to the sound source CPU 10.

Thus, the sound source CPU 10 recognizes that the organ is a timbre control organ based on the signals 00 and 02 sent from the voice recognition section 30, and reads the data of the organ address of the waveform ROM.

On the other hand, if the command is a rhythm control samba, the DSP 32 reads out 20 from the speech recognition dictionary, reads out the rhythm 01 as a control element by referring to the conversion table 33, and then reads out the control item 03. This is read out and sent to the sound source CPU 10.

Thus, the sound source CPU 10 recognizes that it is a rhythm control samba based on the signals 01 and 03 sent from the voice recognition section 30, and determines from the rhythm data samba data address in the data ROM. Read data.

Similarly, in the case of code control Cm7, 27 is read from the voice recognition dictionary, and the code element number 02 and control item 01 are read out from the conversion table 33 and sent to the sound source CPU 10.

As a result, the tone generator CPU 10 recognizes that it is the Cm7 for chord control, and reads from the address of the Cm7 data in the automatic accompaniment chord expansion table.

As described above, the tone generator CPU 10 needs to read data from different tables according to the elements to be controlled. Therefore, the voice recognition unit 30
As in the case of the tone control by the operator, the sound source CPU 10
It is necessary to send control signals for recognizing control elements such as timbre, rhythm, chord, tempo and the like, and control items such as pops and rocks, pianos and trumpets, C # and Am.

For this reason, in the conversion table 33, the control elements controlled by the sound source CPU 10 and the control items in the control elements can be determined based on the identification number of the command determined by collating with the voice recognition dictionary 34. The data which can generate a signal is stored.

Accordingly, when the sound source CPU 10 receives the control signal from the voice recognition unit 30, it first determines the control element,
Subsequently, predetermined control is performed by determining the control item of the control element.

The identification number recognized by the voice recognition unit 30 is sent directly to the sound source CPU 10 and the sound source CPU 1
It is also possible to perform musical tone control by determining a command element or a control item based on the identification number sent from the voice recognition unit 30 at 0.

However, as in the present embodiment, if the voice recognition unit 30 is configured to generate a control signal similar to that in the case of musical tone control by an operator and send it to the sound source CPU unit 10, the electronic musical instrument main body The influence on the configuration is small and the block can be formed, which is convenient when small-scale production of other products is aimed.

The command recognized by voice is not limited to one fixed word, but may be a combination of a plurality of words as long as it can be identified. Also,
The control elements to be voice-controlled are not limited to timbre, rhythm, chord, tempo, and volume, but may include other control elements.

[0077]

As described in detail above, according to the present invention, during performance, a keyword input by voice from a microphone is pattern-recognized, and tone control is performed based on the determined command. Therefore, there is no need to interrupt the performance temporarily during the performance and operate the operator, so that the performance can be performed without interruption and in a natural flow.

Further, by providing a changeover switch for voice control, it can be used as an electronic musical instrument for controlling a musical tone by using a conventional operator, and the convenience is further increased.

[Brief description of the drawings]

FIG. 1 is a diagram illustrating an overall configuration of an electronic musical instrument according to the present invention.

FIG. 2 is a diagram illustrating voice recognition and control operation of a voice control device according to the present invention.

FIG. 3 is a diagram illustrating functions of a conversion table according to the present invention.

[Explanation of symbols]

 Reference Signs List 1 keyboard 2 key press detection unit 3 operation panel 3A voice control switch (voice control switching means) 4 panel scan circuit 5 musical tone signal generation unit 6 waveform memory 7 D / A converter 8 amplifier 9 speaker 10 sound source CPU 11 ROM 12 RAM Reference Signs List 13 microphone (input means) 14 A / D converter 21 command setting section 22 control section 24 tone control table 30 voice recognition section (voice recognition means) 31 voice recognition CPU 32 digital signal processor (DSP) 33 conversion table 34 voice recognition Dictionary 35 Work RAM

──────────────────────────────────────────────────続 き Continued on the front page (51) Int.Cl. ⁶ Identification code FI G10H 1/40 G10H 1/40 1/46 1/46

Claims (3)

[Claims] 1. An electronic musical instrument for controlling a musical tone by operating a manipulator, comprising: input means for voice-inputting a command for musical tone control; and a keyword indicating a musical tone control command from a voice pattern input from the input means. And voice recognition means for converting the signal into a control signal for tone control corresponding to the keyword and transmitting the control signal to the sound source CPU, and a voice pattern corresponding to each command are stored.
A voice recognition dictionary that is collated when the voice recognition unit pattern-recognizes a word that is a keyword for voice control; and a command that is converted by the voice recognition unit into a control signal that can be recognized by the sound source CPU. And a conversion table to be referred to, wherein the sound source CPU can control a musical tone based on a control signal sent from the voice recognition means. 2. A voice control switching means is provided between the voice recognition means and the sound source CPU, and when the voice control switching means is turned on, tone control by voice recognition is possible in addition to a control mode by an operator. The voice control device for an electronic musical instrument according to claim 1, wherein: 3. The voice recognition dictionary and the conversion table have data of one or more control elements of timbre control, rhythm control, chord detection, volume control, and tempo control.
3. The voice control device according to claim 1, wherein voice control of any one or more of a tone color, a rhythm, a chord, a volume, and a tempo is possible.