JP3368912B2 - Musical tone waveform signal generating apparatus and musical tone waveform signal processing method - Google Patents

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a musical tone waveform signal generating apparatus and a musical tone waveform signal processing method capable of generating a variety of musical tone signals.

[0002]

2. Description of the Related Art In conventional sound source devices for electronic musical instruments,
A sampling type sound source device is known in which a musical tone waveform generated by a natural musical instrument is digitally sampled and recorded, and is read out in accordance with a musical tone instruction. In this sound source device, there is known one in which a musical tone waveform is stored for each condition according to the condition directly defining the performance such as touch and pitch, and reproduced based on the condition instructed during the performance. For example, a musical tone waveform when a piano is struck hard and a musical tone waveform when a soft tap is sampled for each range, and these musical tone waveforms are switched and pronounced according to touch information and pitch information during performance,
Alternatively, there is a sound generator which changes the mixing ratio of these in accordance with the touch information.

By the way, it is known that each natural musical instrument has its own directivity. For example, consider a player playing a saxophone. If you listen in front of the performer, you will hear a bright tone. on the other hand. When listening a little behind the performer, the performer's body gets in the way, and high-pitched sound with a strong straight line cannot reach the ears.
You can hear it in a muddy tone. In this way, the timbre differs depending on the presence of the performer in the listening direction, and even if the performer does not exist, the structure of the saxophone itself is complicated, so vibration on the front side, vibration on the side, backside The vibrations in each are different, and the timbre you hear is also slightly different.

Next, consider the case of a piano. In the case of a large musical instrument such as a piano, there is less influence on the sound due to the player's confidence than a saxophone, but since the orientation of the soundboard and reflector is fixed, the tone color differs depending on the orientation of the piano with respect to the listener. . As described above, the characteristic of the generated musical tone depending on the direction of the musical instrument with respect to the listening position is called directivity. It is known that this directivity varies from musical instrument to musical instrument, and that the same musical instrument also varies with pitch. Further, a reference is set for the direction of the musical instrument with respect to the listening position, and the angle from the reference is referred to as a directivity angle.

[0005]

However, in the above-mentioned conventional electronic musical instrument, although the tone color of the musical tone changes depending on the condition (touch, pitch, etc.) that directly defines the performance, the musical tone of the performer and the listener are changed. The timbre did not change depending on the angle or the orientation of the instrument with respect to the listening position.
Therefore, there is a problem that the generated musical sound lacks realism. The present invention has been made in view of the above circumstances, and an object of the present invention is to provide a musical tone signal generating apparatus and a musical tone signal processing method capable of producing a sound taking into consideration the directivity of a sound source.

[0006]

In order to solve the above-mentioned problems, in the invention described in claim 1, in a musical tone waveform signal generator for generating a musical tone waveform signal having a predetermined frequency characteristic, a musical tone source or Directivity angle information output means for outputting directivity angle information representing the directivity angle of the sound image, and a frequency corresponding to the directivity angle information output by the directivity angle information output means.
Frequency characteristic control information output means for outputting numerical characteristic control information
And output by the frequency characteristic control information output means
Control means for controlling the frequency characteristic of the tone waveform signal generated by the tone waveform signal generator based on the frequency characteristic control information .

Further, in the invention described in claim 2,
2. The musical tone waveform signal generating device according to claim 1, wherein the frequency characteristic control information output means corresponds to the directivity angle information output by the directivity angle information output means and the frequency characteristic control information corresponding to pitch information supplied from the outside. Is output. Further, in the invention described in claim 3, the musical tone waveform signal processing for processing the musical tone waveform signal generated
In management apparatus, a recording means for recording in the direction serving as a reference beam angle of sound from the sound source, and to record the sound from emitting sound in a direction at a predetermined angle different from a direction serving as the reference,該収
A filter coefficient calculation for analyzing a difference in frequency characteristics of sounds recorded by the recording means and calculating a filter coefficient for realizing a filter characteristic corresponding to the difference in the analyzed frequency characteristics
Means , a filter coefficient storing means for storing the filter coefficient calculated by the filter coefficient calculating means, an angle information generating means for generating angle information, and an angle information generating means.
Filter coefficient storage means based on the generated angle information
Filter coefficient reading means for reading more filter coefficients,
Filter coefficient read by the filter coefficient reading means
A filter that filters the generated tone signal, whose characteristics are specified by the number.
It is characterized in that it is equipped with a computer . In the invention described in claim 4, the musical tone waveform signal processing according to claim 3
In the device , the pitch information for generating pitch information
Comprising a broadcast generating unit, the generated angle information and characterized by reading the fill <br/> data coefficients from the filter coefficient storage means based on the pitch information.

[0008]

According to the invention of claim 1, the musical tone waveform signal generated by the musical tone waveform signal generator based on the frequency characteristic control information corresponding to the directivity angle information representing the directivity angle of the intended musical tone source or sound image. The frequency characteristic of is controlled to be changed. According to the invention of claim 2, the musical tone waveform signal generator generates the directivity angle information indicating the directivity angle of the musical tone source or the sound image and the frequency characteristic control information corresponding to the pitch information supplied from the outside. The frequency characteristic of the tone waveform signal is changed and controlled. In the invention of claim 3, the recording means emits
The sound from the sound source is recorded in the direction that is the reference of the directivity angle.
The sound source in a direction different from the reference direction by a predetermined angle.
The sound from is recorded. By the filter coefficient calculation means,
Analysis and analysis of differences in the frequency characteristics of recorded sounds
Realizes a filter characteristic corresponding to the difference in frequency characteristics
The filter coefficient to be calculated is calculated. The filter coefficient calculator
The filter coefficient calculated by the stage is stored in the filter coefficient
Stored by the means and generated by the angle information generating means
Filter from the filter coefficient storage means based on the angle information
The coefficient is read by the filter coefficient reading means. This
The filter coefficient read by the filter coefficient reading means
The characteristics of the filter are specified, and the filter generated
The tone signal is filtered. In the invention of claim 4, outgoing
The filter based on the generated angle information and pitch information
The filter coefficient is read from the coefficient storage means,
Tone signal by the filter with the characteristics specified by the data coefficient
Are filtered out .

[0009]

Embodiments of the present invention will be described below with reference to the drawings. FIG. 1 is a block diagram showing the configuration of the first embodiment of the present invention. In the figure, reference numeral 5 denotes a waveform memory, and as shown in FIG. 4, waveform data obtained by sampling a musical sound obtained when a microphone 2 is installed in front of an arbitrary musical instrument, for example, a saxophone 1 is stored. There is. A read control circuit 6 generates address data corresponding to the pitch information PITCH supplied from the outside and supplies it to the waveform memory 5. This address data is
It is updated at intervals according to the pitch information PITCH. That is, the higher the pitch (higher pitch), the address data is updated at larger intervals, and the cycle of the waveform data read from the waveform memory 5 becomes shorter. If known multi-sampling is used in which sampling data is recorded / reproduced separately for each predetermined key area, it is not necessary to increase the address data interval. Furthermore, the known loop processing can be applied to the sampling data of the continuous sound system to reduce the data amount.

Next, 7 is a filter circuit, which performs a filtering process on the waveform data read from the waveform memory 5. The filter coefficient in this filter circuit 7 is
It is supplied from the coefficient control device 8. That is, the filter characteristic of the filter circuit 7 is controlled by the filter coefficient sent from the coefficient control device 8 in real time. The coefficient control device 8 reads the filter coefficient from the coefficient memory 10 according to the pitch information PITCH and the directivity angle information φ. The angle information φ is, as shown in FIG. 4, an angle in the clockwise direction with respect to the front surface of the musical instrument 1 as a sound source.

In the coefficient memory 10, the horizontal axis indicates the pointing angle information φ,
The vertical axis is a table of pitch information PITCH,
At each memory location, a set of filter coefficients Hφ _P (Cφ _P1 , Cφ _P2 , ... Cφ _PN ) (Cφ _PN ) corresponding to the filter order
Is a filter coefficient). The set Hφ _P of these filter coefficients is obtained as follows. First, in the positional relationship where the natural musical instrument and the microphone serve as a reference (0 °), the sound emitted from the natural musical instrument is digitally recorded. At the same time, digital recording is performed with a microphone arranged so as to surround the natural musical instrument at a plurality of positions separated by a predetermined angle.

Fourier transform is performed on the digitally recorded sound data at each position to obtain each frequency characteristic. Then, the difference between the frequency characteristic of the sound of 0 ° and the frequency characteristic at each position is calculated to obtain difference frequency characteristic data. The respective differential frequency characteristic data are inverse Fourier transformed to obtain impulse response data. Then, each coefficient of the FIR filter corresponding to the impulse response is obtained. In this way, a set of filter coefficients is obtained for each predetermined distance. In addition, a natural musical instrument is produced at different pitches, the same analysis as above is performed, and a set of filter coefficients that changes for each pitch (for each PITCH) is obtained.

It is necessary to analyze φ and PITCH at intervals as small as possible and to obtain a set of filter coefficients as closely as possible, but this causes an increase in storage capacity. Therefore, a certain width is analyzed, for example, the directivity angle information φ is analyzed every 30 ° and the pitch information PITCH is analyzed every octave, and a filter coefficient set Hφ is obtained.
_P is obtained in advance, and the interpolation circuit 13 in the coefficient control circuit 8 attempts to perform interpolation. Pitch information PITC
H is supplied by operating a keyboard, a wind instrument type operator, or the like (not shown). Further, the directivity angle information φ is supplied by operating an operator such as a joystick or a trackball (not shown). The read control circuit 6 and the coefficient control circuit 8 each time the information changes.
Is supplied to.

In the above configuration, the pitch information PIT
When CH and directivity angle information φ are supplied, the coefficient control device 8 reads out from the coefficient memory 10 four sets of filter coefficients surrounding adjacent points on the table determined by PITCH and φ and sets them in the interpolation circuit 13. To do. Interpolation circuit 13
Performs a well-known two-dimensional interpolation for each N of the filter coefficients Cφ _PN , and obtains one filter coefficient set from the four filter coefficient sets. The coefficient control device 8 supplies each filter coefficient obtained from the interpolation circuit 13 to the filter circuit 7. As a result, the filter characteristic of the filter circuit 7 becomes
The pitch PITCH corresponds to the directivity φ, and the tone waveform read from the waveform memory 5 is filtered by the filter circuit 7. Therefore, the tone color of the generated musical tone is in line with the directivity. Also, by supplying pitch information and directivity angle information each time there is a change, and by using an interpolation circuit, the filter characteristics can be changed in real time and smoothly,
With a simple operation of operating a joystick or trocar ball, it is possible to generate a musical tone waveform signal equivalent to playing a natural musical instrument while sequentially changing its direction.

In the first embodiment, the filter characteristic is changed according to the directivity angle information φ. However, the waveform data corresponding to the directivity angle information φ is stored separately, and these waveforms are used for the directivity angle. You may comprise so that it may select according to the information (phi). In this case, when simulating a natural instrument, the musical sounds of the natural instrument are sampled from multiple angles,
The sampling result may be stored in the waveform memory. Then, by using the well-known waveform closed interpolation technique, it is possible to obtain a smooth change as in the above-described embodiment.

Although the above-described embodiment is an example in which a musical tone of one tone color is generated, when a plurality of tone colors are generated, the waveform data in the waveform memory 5 and the table in the coefficient memory 10 are generated. May be prepared according to the number of timbres, and these may be selected according to timbre specifying information (a signal output according to a timbre specifying switch, etc.).

Further, the tone waveform is generated by using a sound source such as an FM sound source (frequency modulation sound source) instead of the above-mentioned waveform data reading method, and a sound source program corresponding to each directivity angle is prepared in advance. Good.

Next, the second embodiment will be described. Figure 2
FIG. 7 is a block diagram showing the configuration of the second embodiment. In the figure, 10 is a sound source device configured as in the first embodiment, and 11 is a binaural processor for controlling the position of a sound image. Reference numeral 14 denotes a manipulator boat in which a plurality of manipulators such as a joystick, a jog dial, and a volume slider are arranged. By the operation, directivity angle information φ is sent to the sound source device 10, and information φd and d relating to a sound image position described later are binaural. It is sent to each processor 11. The keyboard 15 gives a tone generation instruction (KON) and pitch information PITCH to the sound source device 10.

The binaural processor 11 will be described. As shown in FIG. 3, the binaural processor 11
Is R based on the information φd corresponding to the angle between the assumed sound image position S and the front of the listener T and the information d corresponding to the distance between the heard position of the listener T and the assumed sound image position S. The volume and volume ratio of the channel and the L channel and the delay time are determined. In addition, the binaural processor 11
Simulates a situation in which sound from an expected sound image position is reflected in an assumed performance space, that is, a hall or a live house, and gives reverberation characteristics corresponding to the sound image position to the R channel and the L channel. . Further, the binaural processor 11 considers the relationship between the arrangement of the speakers and the listening position when the sound is actually emitted, so as to remove the crosstalk components in both the R and L channels.
Perform signal processing on each channel. In this way, the binaural processor 11 faithfully reproduces the position of the sound image not only in the horizontal direction but also in the vertical direction, and operates the joystick or the like provided on the operator board 14 to freely set the sound image. You can move the position.

As described above, according to the second embodiment, by manipulating the manipulator of the manipulator board 14, not only the position of the sound image but also the direction (orientation angle) of the sound image can be freely manipulated. become.

In the second embodiment described above, the filter circuit 7,
Although the filter coefficient memory 10, the coefficient control circuit 8 and the interpolation circuit 13 are arranged in the sound source device 10, if these are arranged in the binaural processor 11, it is possible to obtain what kind of musical tone signal (eg, real time signal) to be input to the binaural processor. Performer's voice input by
Also, the position of the sound image and the direction (directivity angle) of the sound image can be freely manipulated. In this case, it is preferable to use a set of filter coefficients showing a typical directivity change.

[0022]

As described above, according to the present invention,
It is possible to generate a sound with consideration given to the directivity of the sound source, and to generate a musical sound rich in realism.

[Brief description of drawings]

FIG. 1 is a block diagram showing the configuration of an embodiment of the present invention.

FIG. 2 is a block diagram showing a configuration of an application example of the same embodiment.

FIG. 3 is a plan view showing a relationship between a sound source and a listener of the application example shown in FIG.

FIG. 4 is a schematic plan view showing an angle of a sound source.

[Explanation of symbols]

7 ... Filter circuit (control means) 8 ... Coefficient control device (control means), 10 ... Coefficient memory (control means) 14 ... Operator board (direction angle information output means)

Claims (4)

(57) [Claims] 1. A directivity angle information output means for outputting directivity angle information representing a directivity angle of a musical tone source or a sound image in a musical tone waveform signal generator for generating a musical tone waveform signal having a predetermined frequency characteristic, said directivity angle. Frequency characteristic control information output means for outputting frequency characteristic control information corresponding to the directivity angle information outputted by the information output means; and the musical tone waveform based on the frequency characteristic control information outputted by the frequency characteristic control information output means. And a control means for controlling the frequency characteristic of the tone waveform signal generated by the signal generator. 2. The frequency characteristic control information output means outputs frequency characteristic control information corresponding to the directivity angle information output by the directivity angle information output means and pitch information supplied from the outside. The musical tone waveform signal generator according to claim 1. 3. easy waves for processing tone waveform signal generated
In the signal processor , the sound from the sound source is recorded in the direction that becomes the reference of the directivity angle .
The recording means for recording the sound from the sound source in a direction different from the reference direction by a predetermined angle , and the difference in the frequency characteristic of the sound recorded by the recording means are analyzed to determine the difference in the analyzed frequency characteristic. A filter coefficient operation that calculates the filter coefficient that realizes the equivalent filter characteristic.
And calculation means, a filter coefficient storing means for storing a filter coefficient calculated by the filter coefficient calculating means, and the angle information generating means for generating angle information, on the basis of the generated angle information by the angle information generating means
A filter coefficient is read from the filter coefficient storage means.
A filter coefficient reading means, the filter coefficient read out by the filter coefficients read-out means
A filter that filters the generated tone signal, whose characteristics are specified by the number.
Sound waveform signal processing instrumentation, characterized in that it comprises a filter
Place 4. Pitch information for generating pitch information
4. The musical tone waveform signal processing apparatus according to claim 3, further comprising a report generating unit , wherein the filter coefficient is read from the filter coefficient storage unit based on the generated angle information and the pitch information.