JPH0981171A - In-car sound synthesis device on vehicle - Google Patents

Abstract

PROBLEM TO BE SOLVED: To enhance pleasantness of vehicle drive by deciding musical interval and sound volume according to a detected state of driving state based on extracted music scale data. SOLUTION: The vehicle's driving state is detected by inputting signals to CPU 10 from a revolution sensor 18 and a load sensor 20, and a waveform for one segment is generated and the generated waveform data are outputted to a D/A converter 12. In this case, one segment of the sound waveform in each interval of the scale for every musical instrument is stored, and the revolution speed and the load are continuously measured. The tone waveform of a frequency and a tone color corresponding to timewise variable revolution speed and load changing with time can be obtained from the prestored waveforms by means of arithmetic interpolation. The waveforms obtained by the arithmetic interpolation is outputted from the sound generation means 16. Thus, the interval and the sound volume are decided according to the detected driving state based sound on the extracted music scale data.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an in-vehicle sound synthesizing apparatus which generates a musical instrument sound according to a running state of a vehicle in a vehicle interior to improve driving comfort.

[0002]

2. Description of the Related Art Conventionally, a "motor vehicle pseudo exhaust noise generating device" (Japanese Patent Laid-Open No. 1-140199) has been known as a device which artificially generates automobile exhaust noise. The present invention generates white noise, passes it through bandpass filters corresponding to the number of order components, and changes the center frequency of the bandpass filter in synchronization with the engine speed to change the band corresponding to the order component of rotation. The noise is obtained, and these components are variously weighted and superimposed to generate a pseudo exhaust sound.

As another prior art, there is Japanese Patent Laid-Open No. 4-504916, "In-vehicle sound synthesizer". In the present invention, for example, the sound of a racing car or other high-performance car can be heard inside the car.

[0004]

However, the sound output by the "pseudo exhaust sound generator for automobile" or "acoustic synthesizer in vehicle" is an engine sound or a sound similar thereto. It is not possible to provide a comfortable in-vehicle sound to many general drivers who are not car enthusiasts because it is not beyond the range of conventional vehicle sounds.

The present invention has been made to solve the above problems, and an object of the present invention is to generate a comfortable sound representing a driving state of a vehicle in a passenger compartment, thereby improving the comfort of driving the vehicle. Is to improve.

[0006]

According to a first aspect of the present invention for solving the above-mentioned problems, a running state detecting means for detecting a running state of a vehicle is provided in a device for producing a sound in a vehicle traveling by a power source. And musical instrument sound storage means for storing, as musical scale data, waveforms of a plurality of musical tones generated by at least one type of musical instrument, and musical scale data stored in the musical instrument sound storage means are extracted, and based on the musical scale data, The sound processing means for determining the pitch and the volume according to the running state detected by the running state detecting means, and the sound generating means for outputting the sound of the pitch and the volume determined by the acoustic processing means in the vehicle are provided. Is. A prime mover such as an internal combustion engine or an external combustion engine or an electric motor can be used as a power source.

The running state includes the number of revolutions of the power source per unit time, the load, the running vibration, and the like. To determine the pitch, all the pitches of the scale are stored in advance, for example, all the notes of the diatonic scale and the chromatic scale are stored, and the pitch of the generated sound is determined from among them according to the running state, For example, a method of interpolating between chromatic scales so that the pitch changes continuously according to the running state, and further, only two notes of the lowest pitch and the highest pitch are stored in the pitch interpolation, and the pitch between them is A method of calculating by interpolation or the like can be adopted. Also, regarding the determination of the tone color, if a plurality of types of musical instruments are used, the stored tone of the musical instrument is switched step by step according to the running state, tone color interpolation between a plurality of musical instruments is performed, and the tone state is changed according to the running state. It is possible to adopt a method of continuously generating a mixed tone of a plurality of instruments.

According to the second aspect of the present invention, the musical instrument sound storage means stores scale data for a plurality of types of musical instruments, and the acoustic processing means stores a plurality of musical scale data in accordance with the detected running state. The tone color is determined based on the scale data of the musical instrument, and the invention of claim 3 is that the sound processing means interpolates the musical tone data of a plurality of musical instruments according to the detected running state. It is characterized in that interpolation is performed to process the timbre of the scale data.

Further, the invention of claim 4 is characterized in that the number of revolutions of the power source per unit time is detected and the pitch of the scale data is interpolated according to the detected number of revolutions. Is characterized by detecting the load on the power source and performing tone color interpolation according to the detected load.

According to the invention of claim 6, the number of revolutions of the power source per unit time and the load applied to the power source are detected, and the volume is determined according to the detected number of revolutions and the detected load. is there.

Further, according to the invention of claim 7, the pitch of the scale data is interpolated according to the detected number of revolutions, the tone color is interpolated according to the detected load, and the tone speed is interpolated according to the detected number of revolutions and the preload. To determine the volume.

[0012]

The operation and effect of the present invention is stored in advance in the musical instrument sound storage means as a plurality of waveforms of a plurality of tones produced by at least one kind of musical instrument. When the vehicle is running, the running state of the vehicle is detected, and the pitch and volume of the sound to be generated are determined based on the scale data in accordance with the running state, and the determined pitch and volume are output in the vehicle. Therefore, a comfortable musical instrument sound is output according to the running state, and the comfort of driving the vehicle is improved.

According to the invention of claim 2, scale data for a plurality of types of musical instruments are stored, and the tone color is determined based on the scale data of the plurality of musical instruments in accordance with the detected running state. According to the third aspect of the present invention, tone color interpolation is performed by interpolating the musical scale data of a plurality of musical instruments. Therefore, the timbre will change according to the running condition,
A more comfortable sound that indicates the running condition is obtained in the vehicle compartment, and driving comfort is further improved.

According to the invention of claim 4, the pitch changes according to the number of revolutions of the power source, the invention of claim 5 performs the tone color interpolation according to the load, and the invention of claim 6 is the rotation. Since the sound volume is determined according to the number and the load, a more comfortable sound representing the running state can be obtained in the vehicle interior, and driving comfort is further improved.

In the invention of claim 7, the pitch of the scale data is interpolated according to the number of revolutions, the tone color is interpolated according to the load, and the volume is corrected according to the number of revolutions and the load. In addition, a more comfortable sound representing the running state can be obtained in the passenger compartment, and driving comfort is further improved.

[0016]

BEST MODE FOR CARRYING OUT THE INVENTION The present invention will be described below based on a specific embodiment. In this embodiment, an internal combustion engine (hereinafter referred to as an engine) is used as the power source of the vehicle. FIG. 1 shows the overall configuration of the apparatus of this embodiment. The CPU 10 includes a rotation speed sensor 18 that detects the rotation speed of the engine of the vehicle, a load sensor 20 that detects the engine load based on the accelerator pedal opening, a RAM 30, a D / A.
The converter 12 is connected. And the D / A converter 1
An amplifier 14 is connected to the amplifier 2, and a speaker 16 is connected to the amplifier 14. Further, the RAM 30 is formed with a scale data area 301 in which a waveform of one cycle of a tone of each pitch of each musical instrument is stored for each musical instrument.

FIG. 2 is a flow chart showing a main program showing the operation of the CPU 10. Step 10
At 0, signals are input from the rotation speed sensor 18 and the load sensor 20 to detect the running state of the vehicle, a waveform for one segment is generated at step 102, and the generated waveform data is D / A. It is output to the converter 12. This process is repeatedly executed in the cycle of one segment of the waveform. As described above, the method of repeating the processing for one segment and changing the parameter for each segment is generally executed in the processing such as voice synthesis and electronic musical instrument processing. Here, one segment is specifically
As a time length that does not cause a human hearing problem, for example, 10 ms
ec is about 50 msec.

The functional diagram of the above apparatus is shown in FIG. The waveform of one segment of the sound in each pitch of the scale of each musical instrument is stored, and the rotation speed and the load are continuously measured. Then, the frequency corresponding to the rotational speed and the load, which change with time, and the waveform of the timbre sound are obtained by interpolation calculation from the waveforms stored in advance. Then, the waveform obtained by the interpolation calculation is output from the sound generating means.

Regarding the detection of the running state of the vehicle, the value input one cycle before is the value at the start of one segment of the generated waveform, and the value input this time is the value at the end of one segment of the generated waveform. The waveform is generated assuming that the state is continuously changing.

Next, the procedure for generating the waveform for one segment in step 102 of FIG. 2 will be described with reference to FIG. FIG. 4 shows the duration ΔT for one segment.
7 shows how the waveform is generated when the running state represented by the number of revolutions and the load shifts from the segment start point a point to the segment end point b point in the state diagram. The frequency f _s of the generated waveform at point a is determined from the rotational speed at that time, and the frequency f _e of the generated waveform at point b is determined from the rotational speed at that time. Four grid points P _1, P _2, Q _1, Q
₂ is a point among the state points of the waveform data of each musical instrument stored in advance in the scale data area 301 of the RAM 30,
The closest contact including the point b is shown. That is, the grid point P
₁ indicates a pitch of frequency F _{1 on} a harp, grid point P ₂ indicates a pitch of frequency F _{2 on} a harp, grid point Q ₁ indicates a pitch of frequency F _{1 on} a violin, grid point Q ₂ indicates a frequency on a violin The pitch of F ₂ is shown. In this way, the waveform data for one segment at a large number of grid points is prepared in advance and stored in the scale data area 301 of the RAM 30.

The waveform segment for one segment is obtained as follows. A musical instrument, a synthesizer, a natural sound, etc. are sampled, and a waveform segment is generated for each type of sound and for each fundamental frequency, that is, for each pitch of a scale. FIG.
Then, the harp has a soft tone at low load and the violin has a hard tone at high load.

The original sound has a rising portion (attack portion in electronic instrumental terms) and an attenuation portion (decay portion) of the waveform, and the duration thereof reaches several seconds. However, in the present embodiment, the waveform is not used as it is, but as shown in FIG. 5, the waveform is stored after being subjected to looping processing in the electronic musical instrument technology and converted into a periodic waveform of a short time. There is. By doing so, continuous sounds can be synthesized during steady running, and the storage capacity of the RAM 30 can be saved.

As a result of the looping process, the start point and the end point of the waveform are continuous, but the phase of the start point may not be constant depending on the processing method. As a result, as shown in FIG.
A case occurs where the waveform starts with an amplitude of 0 and does not end with an amplitude of 0, that is, the phase of the fundamental frequency waveform does not become π / 2. If the phase of the fundamental wave of each waveform segment is not constant, when performing linear interpolation between waveforms, the waveforms may cancel each other out. Therefore, as shown in FIG. 6, the waveform is cyclically shifted so that the phase of the fundamental wave becomes π / 2.
The method in the figure is to find the phase of the fundamental wave component from the original waveform,
It converts the phase angle into time and shifts that amount.

The four corrugated pieces obtained by the above method are
As shown in FIG. 6, the waveform has a uniform phase and a start point and an end point are continuous. In order to obtain one segment from the point a to the point b, linear interpolation and generation of a sweep waveform are performed from the waveform segments at the four grid points P ₁ , P ₂ , Q ₁ and Q ₂ .

In FIG. 4, the method of generating the waveform when the state point shifts from the point a to the point b is as follows. Note that, for convenience of description, a situation in which the value changes extremely greatly will be described below as an example. The above selected P ₁ is 1000 rpm
m 2, load 1, P ₂ is rotation speed 2000 rpm, load 1, selected Q ₁ is rotation speed 1000 rpm, load 2, and Q ₂ is rotation speed 2000 rpm, load 2. First,
In FIG. 7, from the waveforms at points P ₁ and Q ₁ , R
A waveform is generated by linear interpolation at _one point, that is, at the same load as the load at the point a at a rotation speed of 1000 rpm. This is the internal division of the amplitude of the two waveforms. Next, P ₂ and Q ₂
From the waveform at the point, R ₂ point, that is, the rotation speed 2000 rp
The waveform at the state point of the same load as the load at point b in m is generated by linear interpolation. Then, by using two waveforms at the R ₁ point and the R ₂ point, the rotation speed is from 1000 rpm to the rotation speed is 2000 rpm.
Until the frequency is gradually changed.

Next, a method of gradually changing the frequency will be described with reference to FIG. The waveform c in FIG. 8 corresponds to that in FIG.
The waveform at the load interpolation point R ₁ obtained in (c) of FIG.
The waveform e of is the waveform at the load interpolation point R ₂ obtained in (e) of FIG. Next, changing only the frequency of the waveform c,
A waveform in which the rotation speed is gradually increased from 1000 rpm to 2000 rpm is generated. For this purpose, data at a sampling point different from the sampling point of the waveform c is required. In FIG. 9, black circles are sampling points of the original waveform, and white circles are sampling points of the waveform when the frequency is changed. As the frequency changes, the position of the sampling point of this white circle changes, but the value at that sampling point is obtained by linear interpolation. That is, resampling is performed. This resampling can be performed by linear interpolation as well as zero-order interpolation, although the accuracy is reduced.

If resampling is performed in a short cycle, it is converted into a high frequency, and if resampling is performed in a long cycle, it is converted into a low frequency. By re-sampling the waveform fragments at the frequencies F ₁ and F ₂ , the frequency f
FIG. 10 shows a method of obtaining the resampling period (or resampling time) required to obtain a waveform that changes from _s to the frequency f _e . T _1, t in FIG.
₂ is the starting point when resampling the waveform segment at frequency F ₁ and frequency F ₂ , respectively, and t ₁ ^' _and t ₂ ^' are
These are the end points when resampling the waveform segments at the frequencies F ₁ and F ₂ , respectively.

The equation of step 202 in FIG.

[Number 1] f = f _s + [(f _e -f _s) / ΔT] tau (i-2) is in the interval of [Delta] T, linearly with respect to the sampling point i the frequency from f _s to f _e It is a formula for increasing. τ (i−2) is the time elapsed from the sampling start points t _{1 and} t _{2 up} to the sampling point i. Thus, first, the frequency is increased linearly with time.

The equation of step 204,

## EQU00002 ## Since t _s [i] = t _s [i-1] + τf / F ₁ is the sampling period of the waveform data of frequency F ₁ , τ is proportional to frequency f. It is a formula for increasing. Then, a new sampling time t is obtained by the corrected sampling period τf / F _1.
s [i] is corrected longer than the previous sampling time t _s [i-1].

The determination in step 208 is to determine whether or not the sampling point exceeds the sampling end times t ₁ ^' _and t ₂ ^' . In this way, each sampling time of the waveform data at the frequencies F ₁ and F ₂ is calculated. By calculating the amplitude values of the waveform at these sampling times by interpolation, the waveform when the frequency gradually increases can be obtained.

Further, it is necessary to make continuously generated waveforms continuous, and moreover, it is used for linear interpolation between waveforms.
If the phases of the waveforms of the _two frequencies F ₁ and F ₂ do not match, inconvenience that the waveforms cancel each other out in the linear interpolation between the waveforms. Therefore, the sampling start times t _{1 and} t ₂ must always be set appropriately. It is necessary to select the resampling start point so that the same angle portion from each waveform segment is used as the original waveform for resampling. A calculation method therefor is shown in FIG. Ω represents the phase angle of the starting point. The next sampling start phase Ω is set according to the end time t ₁ ^' of the previously generated waveform (step 306).

In this way, the amplitude and the waveform are changed according to the engine speed and the load, and the amplitude is corrected according to the engine speed and the load. The finally determined waveform is the D / A converter 12
The sound is output from the speaker 16 by being output to the speaker 16.

In the above embodiment, the sound waveform of the musical instrument sound is used as the reference waveform, but a natural sound or a sound waveform of an electronic musical instrument such as a synthesizer may be used.

[Brief description of drawings]

FIG. 1 is a block diagram showing the configuration of an in-vehicle sound synthesizer according to a specific embodiment of the present invention.

FIG. 2 is a flowchart showing a procedure for generating a waveform of an output sound by a CPU.

FIG. 3 is a block diagram showing functions of the apparatus of the embodiment.

FIG. 4 is an explanatory view showing a method for changing a pitch and a timbre according to an engine speed and a load.

FIG. 5 is an explanatory view showing a method for obtaining sound waveform data in each pitch of the musical instrument which is stored in a storage device in advance.

FIG. 6 is an explanatory view showing a method for obtaining waveform data.

FIG. 7 is an explanatory diagram showing a procedure of pitch interpolation and tone color interpolation.

FIG. 8 is an explanatory diagram showing a method of pitch interpolation (frequency sweep).

FIG. 9 is an explanatory diagram showing a method of generating resampling data in pitch interpolation.

FIG. 10 is a flowchart showing a calculation procedure of resampling time.

FIG. 11 is a flowchart showing a calculation procedure for determining a start time for resampling.

[Explanation of symbols]

 10 ... CPU 30 ... RAM 301 ... Scale data area

Claims (7)

[Claims] 1. A device for generating sound in a vehicle that is driven by a power source, wherein: a running state detecting means for detecting a running state of the vehicle; and a plurality of waveforms of notes of a scale emitted by at least one type of musical instrument. Musical instrument sound storage means for storing as data, and scale data stored in the musical instrument sound storage means are extracted, and based on the scale data, in accordance with the running state detected by the running state detecting means, a pitch An in-vehicle sound synthesizer in a vehicle, comprising: an acoustic processing unit that determines a volume and a sound generation unit that outputs a sound of a pitch and a volume determined by the acoustic processing unit in the vehicle. 2. The musical instrument sound storage means stores the scale data for a plurality of types of musical instruments, and the acoustic processing means stores the musical scale data of the plurality of musical instruments in accordance with the detected running state. The in-vehicle sound synthesizer for a vehicle according to claim 1, wherein the tone color is determined based on the scale data. 3. The sound processing means determines a tone color of the scale data by performing tone color interpolation by interpolating tone scale data of a plurality of musical instruments between musical instruments in accordance with the detected running state. An in-vehicle sound synthesizer for a vehicle according to claim 2. 4. The running state detecting means detects the number of revolutions of the power source per unit time, and the acoustic processing means determines the pitch of the scale data according to the detected number of revolutions. The in-vehicle sound synthesizer for a vehicle according to claim 1, wherein the in-vehicle sound synthesizer interpolates. 5. The running state detection means detects a load applied to the power source, and the acoustic processing means performs the tone color interpolation according to the detected load. Item 5. The in-vehicle sound synthesizer in the vehicle according to Item 3. 6. The running state detecting means detects the number of revolutions of the power source per unit time and the load applied to the power source, and the acoustic processing means detects the number of revolutions detected and the detection. The in-vehicle sound synthesizer for a vehicle according to claim 1, wherein the volume is determined according to the applied load. 7. The running state detection means detects the number of rotations of the power source per unit time and the load applied to the power source, and the acoustic processing means responds to the detected number of rotations. 7. The pitch of the scale data is interpolated according to the detected load, the tone color interpolation is performed according to the detected load, and the volume is determined according to the detected rotation speed and the detected load. The in-vehicle sound synthesizer in the vehicle according to item 3.