close

Вход

Забыли?

вход по аккаунту

?

JP2007043295

код для вставкиСкачать
Patent Translate
Powered by EPO and Google
Notice
This translation is machine-generated. It cannot be guaranteed that it is intelligible, accurate,
complete, reliable or fit for specific purposes. Critical decisions, such as commercially relevant or
financial decisions, should not be based on machine-translation output.
DESCRIPTION JP2007043295
An excessive correction of a frequency characteristic causes a phase shift between channels, a
decrease in an SN ratio, and a distortion of an output signal, resulting in noise generation, a
deterioration in sound quality, and a sense of discomfort in hearing. An amplification apparatus
picks up a measurement signal output from a speaker with a microphone, and corrects frequency
characteristics of an output signal output from the speaker to a predetermined target frequency
characteristic based on the collected signal. The signal processing unit compares the measured
frequency characteristic obtained from the collected sound signal with the target frequency
characteristic, and obtains a frequency characteristic correction value for reducing the difference
between the measured frequency characteristic and the target frequency characteristic. An
arithmetic correction value for determining the degree of correction of the frequency
characteristic by the frequency characteristic correction value is determined, and the
characteristic frequency characteristic is corrected using the frequency characteristic correction
value and the arithmetic correction value. [Selected figure] Figure 1
Amplification apparatus and amplitude frequency characteristic adjustment method
[0001]
The present invention amplifies an audio signal and outputs the same, and an amplification
device capable of adjusting the amplitude frequency characteristic of the audio signal output
from the speaker, and adjusts the amplitude frequency characteristic of the audio signal output
from the speaker The present invention relates to an amplitude frequency characteristic
adjustment method that can be performed.
[0002]
08-05-2019
1
In recent years, multi-channel audio reproduction systems such as home theater systems have
become widespread.
In this multi-channel audio reproduction system, the reproduction device reproduces an audio
signal recorded in a recording medium, the audio signal reproduced by the reproduction device
is converted into audio signals of a plurality of channels, amplified, and output to a plurality of
speakers An amplification device is provided.
[0003]
In the amplification device, in order to reproduce an audio signal with high quality sound, the
phase of the audio signal between the channels, the roll-off (cut-off) frequency of the output
signal of each channel, the sound of the entire output signal of each channel It is necessary to set
values of various parameters related to acoustic characteristics such as pressure level, distance
between listening point and speaker, and frequency characteristics of sound pressure level of
output signal of each channel to appropriate values. Conventionally, the listener manually sets
the values of these parameters, but in recent years, there have appeared devices provided with
an automatic sound field correction function that automatically sets the values of these
parameters. Hereinafter, the amplitude frequency characteristic of the audio signal is simply
referred to as "frequency characteristic".
[0004]
The automatic sound field correction function outputs measurement signals from each speaker,
picks up the sound output from each speaker with a microphone installed at the listening point,
and measures the frequency characteristics of the collected audio signal and the entire audio
signal. The sound pressure level, the delay time of the audio signal of each channel, etc. is
compared with the preset frequency characteristics, the sound pressure level, the delay time, etc.,
and the frequency characteristic of the audio signal of each channel to remove the difference.
Correct the sound pressure level of the entire audio signal and the delay time of the audio signal
of each channel. This makes it possible to restore the optimal audio signal at the listening point.
An apparatus provided with such an automatic sound field correction function is disclosed in
Patent Document 1 and Patent Document 2.
08-05-2019
2
[0005]
In the multi-channel audio reproducing apparatus disclosed in Patent Document 1 and the
automatic sound field correction system disclosed in Patent Document 2, the measurement signal
output from each speaker is collected and collected based on the measurement result. The
frequency characteristic of the sounded audio signal is determined, the equalizer coefficient is
automatically determined by calculation so that the frequency characteristic matches the preset
target frequency characteristic, and the equalizer determined by the calculation forms an
equalizer And automatically correct the sound field by outputting an audio signal through the
equalizer.
[0006]
In addition, when the optimum audio signal is restored at the listening point by the automatic
sound field correction function, even when listening around the listening point by calculation
based on each characteristic collected at one listening point Patent Document 3 discloses a
technique for correcting the delay and level of an audio signal output from each speaker so as to
obtain the same reproduction characteristic as when listening at a listening point.
[0007]
Patent Document 1: Japanese Patent Application Publication No. 2000-354300 Japanese Patent
Application Document No. 2001-224092
[0008]
The correction of the frequency characteristic of the audio signal in the multi-channel audio
reproducing apparatus disclosed in Patent Document 1 and the automatic sound field correction
system disclosed in Patent Document 2 is performed by using a target frequency in which the
frequency characteristic obtained by measurement is preset. Correction is performed using
coefficients obtained by calculation so as to match the characteristics.
[0009]
In the measurement of acoustic characteristics by the automatic sound field correction function,
the sound output from the speaker is directly or reflected to the wall of the room and input to the
microphone, so the frequency characteristics of the audio signal picked up by the microphone
are , Peaks with rapidly changing levels (convex parts where the signal level is high) and dips
(concave parts where the signal level is low) or peaks with very large levels and dips are often
08-05-2019
3
generated.
These peaks or dips can be automatically corrected for frequency characteristics by calculation
of automatic sound field correction, but the frequency characteristics obtained by measurement
are corrected so that they match the preset target frequency characteristics. The phase difference
between the channels of the audio signal may occur, the SN (Signal Noise) ratio may decrease, or
distortion may occur in the output signal.
[0010]
Specifically, when correcting a frequency characteristic having a peak and a dip whose level
changes rapidly, that is, a frequency characteristic in which a peak and a dip exist in a narrow
frequency band that human beings can not perceive, Correct the peaks and dips exactly to match.
However, the equalizer used to correct the frequency characteristic in which the frequency band
of peak and dip is narrow has a frequency characteristic having a sharp curve (the signal level
suddenly becomes low or high) in the peak and dip parts, and the peak and dip Since the phase
of the portion of [1] changes extremely, this change in phase causes a shift in the phase
relationship between the channels.
[0011]
In addition, when correcting a frequency characteristic having a peak and a dip having a very
large level, the peak and the dip are accurately corrected to match the target frequency
characteristic in operation.
In the calculation for this correction, a head room must be secured.
The headroom indicates the “degree of margin” that indicates how large a signal can be
handled with respect to the operation level of each audio device, and the level at which the audio
device is distorted and the level of the sound actually output from the audio device And the
difference between For example, if the digital recorder requires a headroom of about 20 dB and
08-05-2019
4
the head amplifier of the mixing console needs about 30 dB, and the maximum signal level is 0
dB, the digital recorder has up to -20 dB as the maximum signal level that can be handled by the
device. In the head amplifier, up to -30 dB is the maximum signal level that can be handled by the
device.
[0012]
Signal processing means (for example, digital signal processor (DSP: Digital Signal Processor))
that digitally performs arithmetic processing for correcting frequency characteristics in order to
secure a head room (for example, 30 dB) in operations for correcting frequency characteristics In
the previous stage of the processing, the analog signal level (gain) of the frequency characteristic
is lowered by a predetermined level, and the signal level of the signal (for example, the analog
signal after digital / analog conversion) after the signal processing means (DSP) Perform
processing to raise the predetermined level. As a result, the head room is secured, and even in
the case of the frequency characteristic having a large level peak, the frequency characteristic
can be corrected without the high level portion being saturated. However, by lowering the signal
level of the frequency characteristic, the dynamic range is narrowed, and a minute level signal is
buried in the noise component, so that the SN ratio is lowered and the sound quality is
deteriorated.
[0013]
Further, since the signal level of the frequency characteristic is lowered to secure the head room,
the signal level needs to be increased in the subsequent stage, so that the circuit scale becomes
large. Further, in the case of frequency characteristics where it is not necessary to originally
secure the head room, it is not necessary to lower the signal level of the frequency
characteristics. In such a case, after the gain of the frequency characteristic is lowered by a
predetermined level to secure the head room, the gain of the frequency characteristic is lowered
in a device that performs processing to increase the gain in the analog circuit in the subsequent
stage. It is necessary to have means for determining whether the gain is not lowered and means
for switching whether to reduce the gain of the frequency characteristic based on the
determination, which complicates the circuit configuration and increases the circuit size.
[0014]
08-05-2019
5
Further, among the speakers, there is a speaker having a frequency characteristic in which a dip
is present in a part of the audible band. Such a dip is a frequency characteristic that the speaker
itself has, and therefore, a signal higher than the signal level of the portion of the dip is not
output. When correction is made to raise the signal level of the portion of the dip by correction
of the frequency characteristic, distortion occurs in the output signal from the speaker due to
clipping of the signal waveform of the frequency at which the signal level is raised.
[0015]
As described above, in the frequency characteristics obtained by measurement, there may be
peaks and dips whose level changes rapidly or peaks and dips of very large levels, or dips in
some bands. When the frequency characteristics are automatically corrected so that the
frequency characteristics obtained by measurement of these peaks or dips are calculated by
automatic sound field correction match the target frequency characteristics, the phase shift
between the channels of the audio signal May cause the sound field to deviate from the normal
position, which may cause a sense of incongruity, or a decrease in S / N ratio may cause noise
components of the audio signal to increase and sound, or distortion or dynamic range of the
output signal may be narrowed. This may cause the sound quality of the audio signal to
deteriorate.
[0016]
Further, in the listening position correction device disclosed in Patent Document 3, the listening
area around the listening point is divided into nine areas based on each characteristic collected at
one listening point, and Assuming that the center of each region is a listening position, the
optimum reproduction condition at each listening position is obtained by calculation, and the
delay and level of the audio signal output from the speaker are corrected. However, if the
listening area around the listening point is too wide or too narrow than the actual listening
position, the optimum reproduction condition can not be obtained at the actual listening position.
[0017]
The present invention reduces the circuit scale without complicating the circuit configuration,
and corrects the frequency characteristic obtained by measurement to match the target
08-05-2019
6
frequency characteristic when correcting the amplitude frequency characteristic of the audio
signal. It is an object of the present invention to provide an amplification device and an
amplitude frequency characteristic adjustment method capable of reducing the a sense of
incongruent sense of discomfort or an increase in noise components and a decrease in sound
quality. Another object of the present invention is to provide an amplification device and an
amplitude frequency characteristic adjustment method capable of obtaining optimum
reproduction conditions even when the listening position is moved.
[0018]
The invention of the present application is an amplification device capable of amplifying and
outputting an audio signal and adjusting the amplitude frequency characteristic of the audio
signal, wherein the amplitude frequency characteristic of a signal output from a speaker
connected to the amplification device is A measurement signal generation unit that generates a
measurement signal to be measured, an output unit that outputs the measurement signal
generated by the measurement signal generation unit, and a microphone that picks up the
measurement signal that is output from a speaker And a signal processing unit that corrects the
amplitude frequency characteristic of an audio signal output from a speaker based on the sound
pickup signal to a predetermined target frequency characteristic to which a sound pickup signal
collected by the sound pickup signal is input And the signal processing unit compares the
measured frequency characteristic obtained from the collected sound signal with the target
frequency characteristic, and determines the minimum signal of the measured frequency
characteristic. The signal level of the measurement frequency characteristic is increased so that
the bell matches the signal level of the target frequency characteristic, and the frequency
characteristic correction value with the smallest difference between the measurement frequency
characteristic and the target frequency characteristic is determined, and the frequency
characteristic An arithmetic correction value for determining the degree of correction of the
frequency characteristic in the correction value is obtained, and the amplitude frequency
characteristic of the audio signal is corrected using the frequency characteristic correction value
and the arithmetic correction value.
[0019]
The invention of the present application is an amplification device capable of amplifying and
outputting an audio signal and adjusting the amplitude frequency characteristic of the audio
signal, wherein the amplitude frequency characteristic of a signal output from a speaker
connected to the amplification device is A measurement signal generation unit that generates a
measurement signal to be measured, an output unit that outputs the measurement signal
generated by the measurement signal generation unit, and a microphone that picks up the
measurement signal that is output from a speaker And a signal processing unit that corrects the
amplitude frequency characteristic of an audio signal output from a speaker based on the sound
08-05-2019
7
pickup signal to a predetermined target frequency characteristic to which a sound pickup signal
collected by the sound pickup signal is input And the signal processing unit is configured to
determine a peak or a peak in the measured frequency characteristic based on the measured
frequency characteristic and the target frequency characteristic. Signal level is detected, and the
signal level of the measurement frequency characteristic is increased so that the signal level of
the minimum dip corresponds to the signal level of the target frequency characteristic, and the
peak or dip is detected. A frequency characteristic correction value composed of the
corresponding filter coefficient and an operation correction value composed of the correction
value to be multiplied by the filter coefficient corresponding to the peak or dip are determined,
and the audio using the frequency characteristic correction value and the operation correction
value The present invention is characterized in that the amplitude frequency characteristic of the
signal is corrected.
[0020]
The invention of the present application is the amplification apparatus described above, wherein
the signal processing unit is configured to set the calculated correction value to an audio based
on the level difference ratio between the peak or dip characteristic and the characteristic based
on the filter coefficient and the characteristic based on the filter coefficient. A correction value
corresponding to the filter coefficient is determined based on a distortion rate of a signal, and an
arithmetic correction value is obtained based on the correction value.
[0021]
The invention of the present application is an amplitude frequency characteristic adjustment
method capable of adjusting an amplitude frequency characteristic of an audio signal,
comprising: generating a measurement signal for measuring an amplitude frequency
characteristic of a signal output from the speaker; Acquiring a measurement frequency
characteristic from a collected sound signal picked up by a microphone which picks up the
measurement signal output from the step, comparing the measured frequency characteristic with
the target frequency characteristic, and the measurement frequency characteristic Increasing the
signal level of the measured frequency characteristic such that the minimum signal level of the
target frequency characteristic matches the signal level of the target frequency characteristic,
and the frequency characteristic correction value having the smallest difference between the
measured frequency characteristic and the target frequency characteristic. Calculating the step of
determining and the degree of correction of the frequency characteristic in the frequency
characteristic correction value Determining a positive value, characterized in that it comprises a
step of correcting the amplitude frequency characteristic of the audio signal using said operation
and correction value and the frequency characteristic compensation value.
08-05-2019
8
[0022]
The invention of the present application is an amplification device capable of amplifying and
outputting an audio signal and adjusting the amplitude frequency characteristic of the audio
signal, wherein the amplitude frequency characteristic of a signal output from a speaker
connected to the amplification device is A measurement signal generation unit that generates a
measurement signal to be measured, an output unit that outputs the measurement signal
generated by the measurement signal generation unit, and a microphone that picks up the
measurement signal that is output from a speaker And a signal processing unit that corrects the
amplitude frequency characteristic of an audio signal output from a speaker based on the sound
pickup signal to a predetermined target frequency characteristic to which a sound pickup signal
collected by the sound pickup signal is input And the signal processing unit is configured to
calculate each of the measurement frequency characteristics obtained from the respective
collected sound signals collected by the microphone at a plurality of measurement positions. The
constant frequency characteristic and the target frequency characteristic are compared, and the
signal level of the measurement frequency characteristic is increased so that the minimum signal
level of the measurement frequency characteristic matches the signal level of the target
frequency characteristic. A frequency characteristic correction value having the smallest
difference between the measured frequency characteristic and the target frequency characteristic
is determined, and an operation correction value is determined and determined to determine the
degree of correction of the frequency characteristic in the frequency characteristic correction
value for each of the determined frequency characteristic correction values. The average value of
the plurality of frequency characteristic correction values and the average value of the plurality
of operation correction values are determined, and the amplitude frequency characteristic of the
audio signal is corrected using the average value of the frequency characteristic correction
values and the average value of the operation correction values. It is characterized by doing.
[0023]
The invention of the present application is an amplitude frequency characteristic adjustment
method capable of adjusting an amplitude frequency characteristic of an audio signal,
comprising: generating a measurement signal for measuring an amplitude frequency
characteristic of a signal output from the speaker; Acquiring a plurality of measurement
frequency characteristics from the collected sound signals collected at a plurality of
measurement positions by a microphone for collecting the measurement signal output from a
plurality of measurement frequency characteristics and each of the plurality of measured
frequency characteristics, Comparing the target frequency characteristics; increasing the signal
level of the measurement frequency characteristics such that the minimum signal level of the
measurement frequency characteristics matches the signal level of the target frequency
characteristics; and the plurality of measurement frequency characteristics The frequency
characteristic correction value with the smallest difference between each measured frequency
characteristic and the target frequency characteristic is calculated. Calculating a correction value
08-05-2019
9
for determining the degree of correction of the frequency characteristic in each frequency
characteristic correction value with respect to the plurality of obtained frequency characteristic
correction values, and calculating a plurality of average values of the plurality of calculated
frequency characteristic correction values The method further comprises the steps of: obtaining
an average value of the operation correction values; and correcting the amplitude frequency
characteristics of the audio signal using the average value of the frequency characteristic
correction values and the average value of the operation correction values. .
[0024]
The invention of the present application is an amplification device capable of amplifying and
outputting an audio signal and adjusting the amplitude frequency characteristic of the audio
signal, wherein the amplitude frequency characteristic of a signal output from a speaker
connected to the amplification device is A measurement signal generation unit that generates a
measurement signal to be measured, an output unit that outputs the measurement signal
generated by the measurement signal generation unit, and a microphone that picks up the
measurement signal that is output from a speaker A pickup signal input unit to which a pickup
signal picked up by the speaker is input, and an arrival time of an audio signal from the speaker
to the microphone is measured based on the pickup signal, and an amplitude frequency of the
audio signal output from the speaker And a signal processing unit that corrects the characteristic
to a predetermined target frequency characteristic, and the signal processing unit is operable to
Each measured frequency characteristic and the target frequency characteristic are compared for
each measured frequency characteristic obtained from each collected sound signal picked up by
the lophon, and the minimum signal level of the measured frequency characteristic matches the
signal level of the target frequency characteristic Thus, the signal level of the measurement
frequency characteristic is increased, and for each measurement frequency characteristic, the
frequency characteristic correction value with the smallest difference between the measurement
frequency characteristic and the target frequency characteristic is determined, and the frequency
characteristic correction is performed for each determined frequency characteristic correction
value. The arithmetic correction value for determining the degree of correction of the frequency
characteristic in the value is determined, and the average value of the plurality of calculated
frequency characteristic correction values and the average value of the plurality of arithmetic
correction values are determined. While correcting the amplitude frequency characteristic of the
audio signal using the average value of the operation correction values, the microphones at a
plurality of measurement positions from each speaker The arrangement of each speaker is
determined based on the arrival time of the audio signal from the speaker to the microphone
obtained from the collected sound signal collected by the phone, and the position of the speaker
and the audio signal of the channel output to the speaker do not match In this case, the output
unit switches to an audio signal of a channel adapted to the speaker.
08-05-2019
10
[0025]
According to the present invention, when correcting the amplitude frequency characteristic of an
audio signal in an amplification device or an amplitude frequency characteristic adjusting method
capable of reducing the circuit scale without complicating the circuit configuration and
correcting the frequency characteristic. By adjusting the frequency characteristic obtained by the
measurement so as to match the target frequency characteristic, it is possible to reduce the sense
of incongruity in hearing, increase in noise, and deterioration in sound quality.
In addition, even when the listening position is moved, it is possible to obtain optimal
reproduction conditions.
[0026]
FIG. 1 is a diagram showing the configuration of a first embodiment of the amplification device of
the present invention.
In FIG. 1, the amplification device includes a signal processing unit 1, an output unit 2, a
measurement signal generation unit 3, a sound collection signal input unit 4, an operation unit 5,
and a control unit 6.
[0027]
The signal processing unit 1 is, for example, a digital signal processor (DSP), and performs signal
processing such as decoding and delay on an audio signal input from an external device.
The decoding process is, for example, a process of converting an audio signal of two channels
into an audio signal of six channels, a process of decompressing a compressed audio signal, and
the like.
The delay processing is processing for adding a delay time to each of the decoded audio signals
of the respective channels in order to give the listener a sense of presence etc. when outputting
08-05-2019
11
audio signals from a plurality of speakers.
[0028]
Under the control of the control unit 6 described later, the signal processing unit 1 listens to the
audio signal in the environment where the audio signal is listened to in the space where the
plurality of speakers are arranged based on the collected signal input from the collected signal
input unit 4. The values of various parameters of acoustic characteristics such as frequency
characteristics, delay time, sound pressure level and the like of audio signals of each channel are
automatically corrected so that an optimum acoustic effect can be obtained at the listening point.
The operation of performing this setting is called automatic sound field correction processing.
[0029]
When correcting the frequency characteristics obtained by measurement to the target frequency
characteristics in the automatic sound field correction processing, the signal processing unit 1
compares these frequency characteristics to obtain a difference, and obtains the frequency
characteristics obtained by measurement. A frequency characteristic correction value to match
the target frequency characteristic is determined.
Then, based on the frequency characteristic correction value, an operation correction value to be
described later that determines the degree of correction by calculation is obtained, and the
frequency characteristic correction is performed using the frequency characteristic correction
value and the operation correction value.
[0030]
Here, the signal processing unit 1 corrects the frequency characteristic of the audio signal by
calculation using the frequency characteristic, the target frequency characteristic, the frequency
characteristic correction value, and the operation correction value obtained by measurement.
08-05-2019
12
In this correction, a parametric equalizer is formed on the calculation program in the signal
processing unit 1, and the correction is performed by changing the frequency of the parametric
equalizer, the value of the signal level, and the like by calculation. The parametric equalizer sets
the frequency of any peak or dip among the peaks or dips in the frequency characteristics of one
band of the graphic equalizer (dividing the audio band and adjusting the level for each band) as
the center frequency Change the frequency characteristics.
[0031]
The signal processing unit 1 includes a memory 1a, and stores values of parameters set by the
automatic sound field correction process in the memory 1a. The audio signal changes the
parameter value so that the listener has the desired frequency characteristic using the operation
unit 5, or until the parameter value is changed by the automatic sound field correction
processing. Signal processing is performed based on the stored value and output.
[0032]
The output unit 2 includes a digital-to-analog conversion unit (not shown) and an amplification
unit (not shown), and converts digital audio signals of a plurality of channels input from the
signal processing unit 1 into analog signals after digital-to-analog conversion. The audio signal is
amplified, and a front left channel speaker (hereinafter referred to as “FL speaker”) connected
to the output unit 2. 7a, front light channel speaker (hereinafter referred to as "FR speaker". 7b,
center channel speaker (hereinafter referred to as "C speaker". 7c, surround left channel speaker
(hereinafter referred to as "SL speaker". 7d, surround light channel speaker (hereinafter referred
to as "SR speaker". 7e, subwoofer channel speaker (hereinafter referred to as "SW speaker". ) 7f
output respectively.
[0033]
The output unit 2 performs control of the control unit 6 so that the listener changes the
frequency characteristics so that the listener has desired frequency characteristics during
automatic sound field correction processing or after automatic sound field correction processing.
When instructed using the operation unit 5, the measurement signal generation unit 3 outputs
the measurement signal generated to the speakers 7a to 7f of each channel.
08-05-2019
13
[0034]
Under the control of the control unit 6, the measurement signal generation unit 3 generates
measurement signals for measuring values of various parameters when performing automatic
correction of frequency characteristics in automatic sound field correction processing.
The measurement signal is a signal having a frequency band wider than the audible band, and is,
for example, impulse, time-stretched pulse, white noise, and the like.
[0035]
The sound pickup signal input unit 4 receives the sound pickup signal input from the
microphone 8 connected to the sound pickup signal input unit 4 in the automatic sound field
correction processing under the control of the control unit 6, and the signal processing unit 1
Output to
[0036]
The operation unit 5 includes an operation button for switching on / off of an automatic sound
field measurement process, an operation button for inputting or changing settings of various
parameters of acoustic characteristics, and the like.
When the listener presses the operation button, the operation unit 5 outputs an instruction signal
corresponding to the pressed operation button to the control unit 6.
[0037]
The control unit 6 comprehensively controls the entire amplification device. The control unit 6
causes the signal processing unit 1 to perform decoding processing, delay processing, and the
like on an audio signal input from an external device, and performs control to output an audio
signal from the output unit 2.
[0038]
08-05-2019
14
When an instruction signal indicating that automatic sound field correction processing is on is
input from the operation unit 5, the control unit 6 causes the measurement signal generation
unit 3 to generate a measurement signal, and the measurement signal is transmitted via the
signal processing unit 1. Thus, control is performed to cause the output unit 2 to output each
speaker. The measurement signal output from the speaker is collected by the microphone 8 and
input to the sound collection signal input unit 4. Then, the control unit 6 causes the signal
processing unit 1 to calculate values of various parameters of acoustic characteristics based on
the collected sound signal input from the collected sound signal input unit 4, and stores the
calculated values in the memory 1a. The value is set as the value of various parameters of the
acoustic characteristic, and the acoustic characteristic is corrected.
[0039]
The automatic sound field correction processing in the amplification device of the first
embodiment will be described. As shown in FIG. 1, in a multi-channel reproduction system such
as a home theater system provided with six speakers 7a to 7f, acoustic characteristics of a room,
frequency characteristics of speakers used, phase characteristics of each channel, and audio
signals from speakers Speaker 7a of each channel at the listening point depending on the
transfer characteristics of the transmission path until the listener listens to the audio signal, the
type and number of speakers used, the installation position of the speakers, the arrangement of
the listening point and each speaker, etc. Because the phase characteristics of the audio signal
from ~ 7f, the roll-off frequency, the sound pressure level of the entire audio signal, the distance
from the listening point to each speaker, and the amplitude frequency characteristics of the audio
signal change, Even if you listen Phase characteristic of a roll-off frequency, sound pressure level,
distance, tone due to the difference in frequency characteristics, sound field, etc. realism different
sounds.
[0040]
An audio signal reproduced by an external device and input to an amplification device is
premised on an appropriate environment for listening to the audio signal. For example, the sound
pressure level of the output signal of each channel is equal, the distance between the speakers 7a
to 7f, the frequency characteristic of each speaker are appropriate, and the user listens in an
environment where speakers of appropriate types are used as a speaker configuration. Because
of the premise, in order to reproduce the audio signal output from the amplification device with
08-05-2019
15
the optimum sound, it is necessary to maintain an appropriate environment.
[0041]
In order to reproduce the audio signal output from the amplification device with optimal sound,
in automatic sound field correction processing, measurement of acoustic characteristics of a
room, characteristics of speakers used, transfer characteristics of transmission path, etc., audio
based on measurement results Make corrections to the signal. The automatic sound field
correction process includes a measurement process and a correction process, and the
measurement process is performed first.
[0042]
The measurement process is a process of measuring the current acoustic characteristic at the
listening point. The speakers 7a to 7f from which audio signals are output are arranged as shown
in FIG. 1, and the microphone 8 for measurement is arranged at a listening point.
[0043]
In the measurement process, the control unit 6 causes the measurement signal generation unit 3
to generate a measurement signal, and causes the signal processing unit 1 to output the
measurement signal from the output unit 2 to the speakers 7a to 7f in order. The measurement
signals output from the speakers 7 a to 7 f are collected by the microphone 8. The sound
collection signal output from the microphone 8 is input to the sound collection signal input unit
4 and is input to the signal processing unit 1. The signal processing unit 1 calculates the impulse
response based on the measurement signal generated by the measurement signal generation unit
3 and the sound collection signal input to the sound collection signal input unit 4. The transfer
characteristic until the signal output from the amplification device is output from each of the
speakers 7a to 7f and the signal is collected by the microphone 8 is the impulse response
obtained here.
[0044]
Based on this impulse response, the sound pressure level and peak level of the output signal of
each channel are obtained by calculation, and FFT (Fast Fourier Transform) analysis is performed
08-05-2019
16
to obtain a speaker configuration (presence or absence of a speaker, speaker size), and between
speakers The phase relationship, the distance between the speakers, the distance from the
listening point to each speaker, the sound pressure level and the frequency characteristic of the
output signal from each speaker can be obtained.
[0045]
Then, the process shifts to correction processing.
The correction process corrects various parameters based on the measurement result obtained
by the measurement process.
[0046]
The signal processing unit 1 sets the number of speakers based on the presence or absence of
the speakers, corrects the delay time between the channels based on the distance between the
speakers or the distance from the listening point to each speaker based on the measurement
result obtained in the measurement process. Correct the difference in sound pressure level of the
output signal between each channel based on the sound pressure level of the output signal of
each speaker, correct the frequency characteristic of the output signal of each channel based on
the frequency characteristic of the output signal of each speaker, etc. The coefficient for this is
determined by calculation.
[0047]
The coefficients obtained by calculation are stored in the memory 1a, and when outputting an
audio signal input from an external device, correction of the frequency characteristic of the audio
signal is performed using the coefficients stored in the memory 1a Signal processing such as
correction of sound pressure level, delay of output time, etc. are performed and output.
[0048]
The amplification apparatus calculates various values of the parameters of the signal processing
unit 1 (size of speaker, number of speakers, distance between speakers, distance from listening
point to each speaker, delay time of each channel, sound of each channel The display of the
amplifier (not shown) so that the listener can confirm the pressure level, the frequency
characteristics of each channel, etc.) and what settings are set as a result of the automatic sound
field correction. Alternatively, it is displayed on a monitor (not shown) connected to the
08-05-2019
17
amplification device, and the automatic sound field correction processing is completed.
[0049]
Next, correction of frequency characteristics in automatic sound field correction processing will
be described.
The signal processing unit 1 is provided in advance with information indicative of a plurality of
target frequency characteristics (values indicative of target frequency characteristics), and these
target frequency characteristics are determined by the listener before the measurement
processing of the automatic sound field correction processing. It is possible to select using the
operation unit 5.
In the correction processing of the automatic sound field correction processing, a frequency
characteristic correction value for making the frequency characteristic obtained from the
measurement result match the target frequency characteristic is determined, and the operation
correction value is determined based on the frequency characteristic correction value. The
frequency characteristic is corrected by calculation using the value.
Then, the signal processing unit 1 outputs the audio signal of the corrected frequency
characteristic.
[0050]
FIG. 2 is a view for explaining correction processing of automatic sound field correction
processing in the amplification device of the first embodiment. FIG. 3 is a diagram for explaining
an operation relating to the filter coefficient Q in the amplification device of the first
embodiment. FIG. 4 is a diagram for explaining the operation regarding the operation correction
value R in the amplification device of the first embodiment. In the automatic sound field
correction process, the control unit 5 causes the measurement signal generation unit 3 to
generate a measurement signal, and measures the frequency characteristics of each speaker. As a
result of measurement, it is assumed that the frequency characteristic A shown in FIG. 2A is
obtained.
08-05-2019
18
[0051]
The frequency characteristic A in FIG. 2A is a frequency characteristic obtained by smoothing the
frequency characteristic obtained by measurement by the signal processing unit 1 for each
octave band such as 1/3 octave or 1/6 octave. This is because the frequency characteristics
obtained by the measurement include frequency characteristics including large peaks and
complex peaks and dips due to the frequency characteristics of the speaker, the acoustic
characteristics of the room, the arrangement of the speaker, and the influence of reflection from
an object in the room. Become.
[0052]
Assigning a parametric equalizer to each peak and dip in order to correct all the peaks and dips
of such frequency characteristics is difficult due to the complexity and time required for
computation. Therefore, the frequency characteristic obtained by the measurement is smoothed,
and the frequency characteristic is corrected for the smoothed frequency characteristic A.
[0053]
FIG. 2B is an example of the target frequency characteristic T. The correction of the frequency
characteristic in the automatic sound field correction processing is performed such that the
frequency characteristic A shown in FIG. 2A approaches the target frequency characteristic T
shown in FIG. 2B. Here, the target frequency characteristic T shown in FIG. 2B is, for example, a
signal level of a level (-30 dB) obtained by lowering a predetermined signal level to secure the
headroom with respect to the maximum signal level (0 dB). Have.
[0054]
When the user selects a device of the audio signal source to be reproduced using the operation
unit 5, the control unit 6 automatically determines a value (signal level) for securing a head room
corresponding to the selected device. . The value (signal level) for securing the headroom is
stored in advance in the memory (not shown) together with the type of the device, and the
control unit 6 refers to the information stored in the memory to secure the headroom. Determine
08-05-2019
19
(signal level). The signal level of the target frequency characteristic T is automatically set by the
control unit 6 to a signal level corresponding to the head room according to the connected
device.
[0055]
The control unit 6 controls the signal processing unit 1 and first detects the frequency of the
portion with the lowest signal level, which is the dip portion of the frequency characteristic in the
frequency characteristic A shown in FIG. Let f (n) be the frequency. Also, the signal level x (n) at
the reference frequency f (n) is detected. Next, in the target frequency characteristic T shown in
FIG. 2B, the signal level x ′ (n) at the reference frequency f (n) is detected.
[0056]
The reference frequency f (n) is a frequency in a frequency band (e.g., 500 Hz to 5 kHz) that is
most sensitive to changes in signal level in terms of hearing included in the band of human voice.
That is, in the correction of the frequency characteristic by the conventional automatic sound
field correction processing, the frequency characteristic obtained by the measurement is made to
match the target frequency characteristic exactly, so the sense of incongruity in the sense of
hearing by the phase shift, the increase of noise, the deterioration of the sound quality It occurs.
[0057]
In the automatic sound field correction processing of the present embodiment, the target
frequency characteristic is the frequency characteristic obtained by measurement in a certain
frequency band in order to reduce the sense of incongruity on hearing, the increase of noise
components, and the deterioration of sound quality Correction is made to match accurately, and
the degree of correction in the peak and dip areas is lowered in other frequency bands, and
measurement is performed to reduce the sense of incongruous sense of hearing due to phase
shift, increase in noise components, and deterioration in sound quality. The obtained frequency
characteristic is corrected to be close to the target frequency characteristic.
[0058]
In the correction of the frequency characteristic of the first embodiment, the frequency
08-05-2019
20
characteristic obtained by measurement and the target frequency characteristic do not match
exactly in the portion of the peak or dip where the difference between the frequency
characteristic obtained by measurement and the target frequency characteristic is large. Part
occurs.
If the part where the frequency characteristic obtained by this measurement and the target
frequency characteristic do not match exactly is the frequency of the frequency band most
sensitive to changes in the signal level, the frequency characteristic in that part is appreciably
corrected. Not to be perceived by the listener.
[0059]
For this reason, in the correction of the frequency characteristic in the present embodiment, the
reference frequency for correcting the frequency characteristic is set to the frequency of the
frequency band most sensitive to changes in the signal level according to human auditory sense.
Since the obtained frequency characteristic matches the target frequency characteristic, the
listener is not perceived as if the frequency characteristic is not much corrected.
[0060]
The signal level x (n) of the reference frequency f (n) in the frequency characteristic A shown in
FIG. 2 (a) is the signal level x '(n) of the reference frequency f (n) in the target frequency
characteristic T shown in FIG. The overall signal level of frequency characteristic A is increased
to match n).
As a result, the frequency characteristic A shown in FIG. 2 (a) becomes the frequency
characteristic A 'shown in FIG. 2 (c).
[0061]
As shown in FIG. 2C, the frequency characteristic A ′ is higher than the signal level of the target
frequency characteristic T, so the frequency characteristic used when correcting the frequency
characteristic A described later is always the same as the target frequency characteristic T. The
negative (minus side) frequency characteristic is obtained, and the corrected frequency
characteristic is also lower than the signal level of the target frequency characteristic T. That is,
08-05-2019
21
the corrected frequency characteristic is lower than the signal level (−30 dB in the present
embodiment) for securing the head room. As a result, the frequency characteristic after
correction is corrected such that the head room is secured and the frequency characteristic close
to the target frequency characteristic T.
[0062]
Next, in the frequency characteristic A ′ shown in FIG. 2C, the signal level of the target
frequency characteristic T and the signal level of the frequency characteristic A ′ are compared
for each frequency, and a signal higher than the signal level of the target frequency characteristic
T The absolute value of the signal level difference g with respect to the signal level of the target
frequency characteristic T in the portion (peak portion) of the level is detected. In FIG. 2C, the
absolute value of the signal level difference g1 at the frequency fc1 is the largest, then the
absolute value of the signal level difference g2 at the frequency fc2 is the largest, and the
absolute values of the signal level difference g3 at the frequency fc3 are It is assumed that the
smallest is detected.
[0063]
When the peak portion of the signal level difference g1 having the largest absolute value of the
signal level difference g is detected, the frequency fc1 of the signal level difference g1 is set as
the center frequency of the parametric equalizer, and the signal level difference g1 is set in the
parametric equalizer Set as the level of frequency characteristic correction.
[0064]
Then, as shown in FIG. 2D, for the peak portion of the characteristic P1 of the signal level
difference g1 at the frequency fc1, the filter coefficient Q and the correction value R for bringing
the peak portion of the characteristic P1 closer to the target frequency characteristic T Ask.
The filter coefficient Q is ideally a value that can obtain a characteristic symmetrical to the peak
portion of the characteristic P1. Further, the correction value R is a value for determining the
degree of correction of the frequency characteristic at the peak portion of the characteristic P1
by the parametric equalizer using the filter coefficient Q.
08-05-2019
22
[0065]
As for how to obtain the filter coefficient Q, the signal processing unit 1 is previously provided
with a plurality (for example, 10) of filter candidate coefficients q (n) as a table, and the signal
level difference g1 and the frequency fc1 of the peak portion of the characteristic P1. A filter is
configured based on each of the filter candidate coefficients q (n) to form a parametric equalizer
using the filter. A difference is calculated by comparing the signal level of the frequency
characteristic and the characteristic P1 of this parametric equalizer for each frequency in the
frequency band of the peak portion of the characteristic P1, and the filter candidate coefficient q
(n ) Is the filter coefficient Q1 of the peak portion of the characteristic P1.
[0066]
Specifically, as shown in FIG. 3A, a parametric equalizer is formed for each of the peak portions
of the characteristic P1 by using a plurality of filter candidate coefficients qa to qj provided in
advance by calculation. For each filter candidate coefficient qa to qj, compare the frequency
characteristic of the parametric equalizer and the characteristic P1 for each frequency in the
frequency band of the peak part of the characteristic P1, and sum the level difference of each
frequency (difference Calculate). As a result, for each of the filter candidate coefficients qa to qj,
the difference in signal level between the frequency characteristic of the parametric equalizer
using the filter candidate coefficients qa to qj and the characteristic P1 is obtained.
[0067]
The difference in signal level between the frequency characteristic of the parametric equalizer
using the filter candidate coefficients qa to qj and the characteristic P1 is as shown in FIG. 3B,
and the frequency characteristic and the characteristic P1 determined by the parametric
equalizer are obtained. And the filter candidate coefficient qg with the smallest difference
between them is the filter coefficient Qg.
[0068]
Next, for the obtained filter coefficient Qg, a correction value R for determining the degree of
correction of the frequency characteristic by the parametric equalizer using the filter coefficient
Qg is obtained.
08-05-2019
23
The degree of correction of the frequency characteristic in the dip section changes with the peak
level of the characteristic based on the filter coefficient Qg. For example, the lower the peak level
of the characteristic based on the filter coefficient Qg, the lower the degree of correction, and the
higher the peak level of the characteristic based on the filter coefficient Qg the higher the degree
of correction. The degree of correction is determined by multiplying the filter coefficient Qg by
the correction value R.
[0069]
As the correction value R, one correction value R is selected from among a plurality of correction
candidate values r (n) obtained based on the signal level difference g1 of the peak portion of the
characteristic P1. This selection method will be described later.
[0070]
Each correction candidate value r (n) is obtained by equally dividing the signal level difference g1
of the peak portion of the characteristic P1 by a predetermined number (for example, 10), and
each equally divided level becomes a peak level. It shows characteristics based on the filter
coefficient Qg. For example, the characteristics of the filter coefficient Qg based on the correction
candidate values ra to rj become characteristics as shown in FIG. 4A, respectively, and are
obtained by multiplying the filter coefficient Qg by the calculation correction candidate values ra
to rj.
[0071]
Possible values of the correction candidate value r (n) are, for example, 0.1 to 1. When the
correction value R selected from the plurality of correction candidate values r (n) is 0.1, the
frequency characteristic is corrected by 10% by the parametric equalizer using the filter
coefficient Qg (correction 10%: FIG. In a), when the correction value R selected from the
correction candidate value ra) and the plurality of correction candidate values r (n) is 1, the
frequency characteristic is corrected 100% by the parametric equalizer using the filter coefficient
Qg (Correction 100%: correction candidate value rj in FIG. 4A). When the correction value R is 0,
no correction of the frequency characteristic is performed by the parametric equalizer using the
filter coefficient Qg (correction 0%), so the value of 0 is excluded from the correction candidate
08-05-2019
24
value r (n).
[0072]
For example, when the center frequency fc (n) of the peak portion is 500 Hz and the signal level
(peak level) of the frequency 500 Hz is 10 dB lower than the signal level of the target frequency
characteristic, the frequency characteristic by the parametric equalizer using the filter coefficient
Qg , Constitute a correction equalizer that amplifies the signal level at a frequency of 500 Hz. In
this case, if the correction candidate value r (n) is +1 dB correction, r = 0.1 (correction 10%), if +2
dB correction r = 0.2 (correction 20%),... +10 dB correction In this case, r = 1 (correction 100%).
From among these values, an appropriate correction candidate value r (n) is selected as a
correction value R to be used for correction of the frequency characteristic of the peak portion of
the characteristic P1 by calculation described later.
[0073]
The parametric equalizer of the characteristic (FIG. 4 (a)) of the filter coefficient Qg based on
each correction candidate value ra to rj is formed, and the level difference between the
characteristic P1 and the characteristic of the filter coefficient Qg based on each correction
candidate value ra to rj The distortion rate of the audio signal in the characteristics of the rate
and the filter coefficient Qg based on the correction candidate values ra to rj is determined.
[0074]
First, the level difference ratio between the characteristic P1 and the characteristic of the filter
coefficient Qg based on the correction candidate values ra to rj will be described.
The level difference rate calculates the sum Xa of the absolute values of the signal levels of the
respective frequencies of the characteristic P1, and the sum Xb of the absolute values of the
signal levels of the respective frequencies of the characteristics of the filter coefficient Qg based
on the respective correction candidate values ra to rj. The level difference rate is calculated by
dividing the total sum Xb by the total sum Xa.
[0075]
08-05-2019
25
The sum Xa uses the signal level of the target frequency characteristic T as a reference value,
determines the signal level of the characteristic P1 with respect to the reference value at each
frequency within the frequency band of the peak part of the characteristic P1, and obtains the
absolute value of the obtained signal level Calculated as the total sum Xb. In addition, with
respect to the characteristics of the filter coefficient Qg based on the correction candidate values
ra to rj, the sum Xb takes the signal level of the portion where the signal level in the
characteristics of the filter coefficient Qg is flat as a reference value. At each frequency in the
frequency band corresponding to the band, the signal level of the characteristics of the filter
coefficient Qg based on the correction candidate values ra to rj with respect to the reference
value is determined, and the absolute value of the determined signal level is calculated as the
sum Xa.
[0076]
The level difference rate is, for example, a characteristic diagram as shown in FIG. In the case of
the correction candidate value ra, as shown in FIG. 4A, since the peak level of the characteristic
of the filter coefficient Qg is low, the total sum Xb in the characteristic of the filter coefficient Qg
based on the correction candidate value ra decreases. As a result, in the case of the correction
candidate value ra, the level difference ratio between the sum Xb and the sum Xa of the
characteristic P1 becomes high as shown in FIG. 4 (b). That is, the characteristic of the filter
coefficient Qg based on the correction candidate value ra indicates that the degree of correction
of the frequency characteristic is low.
[0077]
Further, in the case of the correction candidate value rj, as shown in FIG. 4A, the peak level of the
characteristic of the filter coefficient Qg is high, so the total sum Xb in the characteristic of the
filter coefficient Qg based on the correction candidate value rj becomes large. As a result, in the
case of the correction candidate value rj, the level difference ratio between the total sum Xb and
the total sum Xa of the characteristic P1 becomes low as shown in FIG. 4 (b). That is, the
characteristic of the filter coefficient Qg based on the correction candidate value ra indicates that
the degree of correction of the frequency characteristic is high.
[0078]
Then, as shown in FIG. 4B, the level difference ratio in the characteristic of the filter coefficient
08-05-2019
26
Qg based on the characteristic P1 and each of the correction candidate values ra to rj is
substantially in the direction from the correction candidate value ra to the correction candidate
value rj. It becomes the characteristic which becomes low linearly.
[0079]
Thus, as shown in FIG. 4B, the level difference ratio between the characteristic P1 and the
characteristic of the filter coefficient Qg based on the correction candidate values ra to rj is the
characteristic of the filter coefficient Qg based on the correction candidate values ra to rj. The
higher the deviation from the symmetrical characteristic of the characteristic P1, the lower the
level difference rate, the closer the characteristic of the filter coefficient Qg based on the
correction candidate values ra to rj is closer to the symmetrical characteristic of the
characteristic P1.
[0080]
Next, the distortion rate of the audio signal in the characteristics of the filter coefficient Qg based
on the correction candidate values ra to rj will be described.
The distortion rate has a peak level of each of the correction candidate values ra to rj for each of
the correction candidate values ra to rj, and the sine wave of the center frequency of the peak
portion in the characteristic P1 (fc1 shown in FIG. The sound is collected by the microphone, and
the geometric mean value of the sum of the effective values of the harmonic components of the
collected signal is determined, and the ratio of the geometric mean value to the effective value of
the fundamental wave of the output signal Ask.
[0081]
In the frequency characteristic having the peak portion of the characteristic P1, if the correction
of the frequency characteristic is performed to raise the level of the audio signal output from the
speaker above the level of the peak portion, the signal waveform of the frequency with the
increased level is clipped. , Distortion (harmonic component) occurs in the output signal from the
speaker.
For each of the correction candidate values ra to rj, the distortion rate is determined using the
08-05-2019
27
signal output from the speaker and the signal collected by the microphone.
[0082]
Specifically, a sine wave having a peak level of the correction candidate value ra is output from
the speaker, the sound is picked up by the microphone, and a geometric average value of the sum
of effective values of harmonic components of the collected signal is obtained. The ratio of the
geometric mean value to the effective value of the fundamental wave of the output signal is
determined. Next, a sine wave having a peak level of the correction candidate value rb is output
from the speaker, the sound is picked up by the microphone, and a geometric mean value of the
sum of effective values of harmonic components of the collected signal is determined. The ratio
of the average value to the effective value of the fundamental wave of the output signal is
determined. This operation is repeated to obtain distortion rates in each of the correction
candidate values ra to rj.
[0083]
Then, the distortion rate of the audio signal is, for example, a characteristic diagram as shown in
FIG. Note that, in FIG. 4C, the distortion rate on the vertical axis is expressed as logarithm so that
a value of 1% or less can be visually recognized in describing the distortion rate threshold
described later. In FIG. 4C, in the case of the correction candidate values ra to rd, the distortion
rate is low, the distortion rate gradually increases from the correction candidate value re, and the
distortion rate becomes the highest in the correction candidate value rj.
[0084]
The distortion rate is lower than a predetermined distortion rate (distortion rate threshold) from
the distortion rate of the audio signal in the characteristics of the filter coefficient Qg based on
the correction candidate values ra to rj thus obtained and the above-mentioned level difference
rate A correction candidate value r (n) which is a rate and has the lowest level difference rate is
determined. The predetermined distortion rate is, for example, 0.01%, which is the distortion rate
of a general audio device.
[0085]
08-05-2019
28
For example, in the case of the characteristic chart shown in FIG. 4C, the correction candidate
value rg is a value whose distortion rate is lower than the distortion rate threshold (0.01%) and
whose level difference rate is low. That is, the degree of correction of the frequency characteristic
of the dip portion of the characteristic P1 is high, and the distortion ratio of the audio signal is a
low correction candidate value rg, and the correction candidate value rg is set as the correction
value R1.
[0086]
In the first embodiment, the distortion ratio of the audio signal in the characteristics of the level
difference between the characteristic P1 and the characteristics of the filter coefficient Qg based
on the correction candidate values ra to rj and the characteristics of the filter coefficient Qg
based on the correction candidate values ra to rj First, the level difference rate and the distortion
rate are obtained for the correction candidate value ra, and then the values of the correction
candidate values ra to rj are sequentially changed to obtain the level difference rate and the
distortion rate for the correction candidate value rb The level difference rate and the distortion
rate are determined for, but the operation is stopped when the correction candidate value r (n)
whose distortion rate exceeds the distortion rate threshold is obtained, and the correction value R
is determined by the operation result up to that point You may By this, it is possible to shorten
the calculation time required to obtain the correction value R.
[0087]
As a result of these calculations, for the peak portion of the characteristic P1 of the center
frequency fc1 with the signal level difference g1, the filter coefficient Qg and the correction value
R1 used for the parametric equalizer for correcting the peak portion are determined. If a
parametric equalizer is formed based on these values, the characteristic S1 shown in FIG. 2E is
obtained as the characteristic for correcting the peak portion. If the peak part of characteristic P1
of center frequency fc1 is corrected with the signal level difference g1 by the said parametric
equalizer, it will become as shown in FIG. 2 (f).
[0088]
08-05-2019
29
After the filter coefficient Qg for correcting the peak portion of the characteristic P1 and the
correction value R1 are determined, next, in the frequency characteristic A ′ shown in FIG. 2C,
the signal level difference where the absolute value of the signal level difference g is maximum
For the peak portion of g2, the filter coefficient Q and the correction value R2 to be
characteristics S2 for correcting the peak portion of the signal level difference g2 are determined
by the same calculation as described above. By repeating this process, the filter coefficient Q and
the correction value R for each peak portion of the frequency characteristic A 'are obtained.
[0089]
The value indicating the characteristic consisting of all the filter coefficients Q determined for
each peak of the frequency characteristic A 'in this manner is the frequency characteristic
correction value. Further, a value consisting of all the correction values R obtained for each peak
of the frequency characteristic A 'is an operation correction value to be multiplied by the
frequency characteristic correction value. A parametric equalizer having a frequency
characteristic C shown in FIG. 2G is formed using these frequency characteristic correction
values and operation correction values, and the frequency characteristic of the audio signal is
corrected using the parametric equalizer.
[0090]
When the frequency characteristic A shown in FIG. 2A is corrected using the frequency
characteristic C shown in FIG. 2G, the frequency characteristic D shown in FIG. 2H is obtained. As
in the frequency characteristic D shown in FIG. 2H, the frequency characteristic is corrected to
match the target frequency characteristic T, and a head room is secured.
[0091]
As described above, when correcting the peak or dip portion in the correction of the frequency
characteristic, the signal level at the reference frequency is the target frequency characteristic
with the frequency of the minimum value of the signal level of the acquired frequency
characteristic as the reference frequency. The signal level of the frequency characteristic is
increased to match the signal level, and the frequency characteristic is corrected using the
frequency characteristic correction value and the operation correction value in which the
difference between the acquired frequency characteristic and the target frequency characteristic
08-05-2019
30
decreases. Therefore, it is possible to suppress the phase shift of the audio signal generated by
correcting the frequency characteristic so that the frequency characteristic obtained by the
measurement matches the target frequency characteristic, the reduction of the SN ratio, and the
generation of the distortion of the output signal. It is possible to reduce the above-mentioned
sense of incongruity, noise generation and sound quality deterioration. In addition, processing for
increasing the signal level of the frequency characteristic for securing the headroom at the stage
prior to the correction processing for the frequency characteristic and processing for correction
of the frequency characteristic for the digital analog converted signal, the headroom is ensured.
In order to increase the signal level, it is not necessary to perform processing to decrease the
signal level.
[0092]
In the first embodiment described above, the correction of the frequency characteristic is
performed in the automatic sound field correction processing, however, it is also performed
when the listener manually adjusts the frequency characteristic after the automatic sound field
correction processing. May be
[0093]
In the first embodiment, the correction value R is the level difference ratio between the
characteristic P1 and the characteristic of the filter coefficient Qg based on the correction
candidate values ra to rj, and the characteristic of the filter coefficient Qg based on the correction
candidate values ra to rj. Automatically determined on the basis of the distortion rate of the audio
signal at the time the listener listens to the audio signal output from the speaker, and using the
operation unit 5 determines that distortion does not occur in the aurally audible audio signal The
corrected value may be determined as the correction value R.
[0094]
Further, in the first embodiment, the amplification apparatus capable of amplifying and
outputting the audio signal and adjusting the frequency characteristic of the audio signal output
from the speaker has been described, but the frequency characteristic of the audio signal input
from the outside is described. It can also be applied to an equalizer capable of adjusting.
[0095]
In the first embodiment described above, the measurement point at which the output signal from
the speaker is measured in the automatic sound field correction processing is one listening point,
08-05-2019
31
but the present invention is not limited thereto. A field correction process may be performed.
In the automatic sound field correction process described above, the measured point is the
optimum listening point, but if the listener listens at a position slightly offset from the
measurement position, or if there are multiple listeners, You will listen off the optimal listening
point.
Next, as the second embodiment, a process of performing measurement at a plurality of locations
in the automatic sound field correction process of the first embodiment described above and
performing sound field correction based on the measurement result will be described.
[0096]
The amplifying device in the second embodiment is the same as the configuration shown in FIG.
1, and thus the detailed description will be omitted.
The second embodiment differs from the amplification device of the first embodiment in the
processing of the control unit and the signal processing unit.
[0097]
In the second embodiment, first, frequency characteristics are measured at a listening point (a
position substantially at the center of a plurality of speakers) when there is one listener, and a
frequency characteristic correction value and an operation correction value are calculated. The
frequency characteristic C for correction shown in FIG. 2 (g) is obtained using the frequency
characteristic correction value and the operation correction value.
[0098]
In the first embodiment, since only one measurement is performed for one listener, only one
frequency characteristic C for correction can be obtained. Since this characteristic is a
characteristic in the case where the listener is limited to the place where the measurement
08-05-2019
32
microphone is installed, strictly speaking, the correction is not reflected at the place where the
listener moves even a little from the place. In addition, if the frequency characteristic is corrected
by forming a parametric equalizer based on the frequency characteristic C for correction
obtained at the measurement position, the correction effect of the frequency characteristic is
weakened only at a place away from the measurement position. On the contrary, the sound
quality may be deteriorated audibly by correction.
[0099]
In the second embodiment, in addition to the listening points, the measuring points are increased
to several places near the listening points. In this embodiment, four measurement points are
added in addition to the listening point, and measurement is performed at a total of five places.
This makes it possible to perform acoustic correction in consideration of a relatively wide range
near the listening point by grasping the feature of the acoustic characteristic near the listening
point and performing automatic sound field correction based on the characteristic.
[0100]
The frequency characteristics between the plurality of speakers and the microphone obtained by
one measurement point are different even at a position slightly moved from the measurement
point. However, when a plurality of frequency characteristics measured at a plurality of locations
near the measurement point are statistically analyzed, there are portions similar to the frequency
characteristics. This similar part is a characteristic not influenced by the measurement position,
that is, the frequency characteristic of the speaker itself and the acoustic characteristic at the
measurement point deviated from the listening point.
[0101]
In the second embodiment, the frequency characteristics for correction of five corrections
obtained by measurement and calculation are averaged for each frequency as the simplest
statistical analysis method in consideration of reduction of calculation time and downsizing of the
amplification device. This averaging process makes the frequency characteristics unique to each
measurement point inconspicuous. In addition, it is possible to extract by averaging the
frequency characteristics common to each measurement point, and to extract only the inherent
characteristics of the plurality of speakers and the general spatial characteristics of the room
08-05-2019
33
used for listening.
[0102]
FIG. 5 is a diagram for explaining measurement of frequency characteristics in automatic sound
field correction processing of the amplification device of the second embodiment of the present
invention. FIG. 6 is a diagram for explaining the processing for obtaining the correction
frequency characteristic in the automatic sound field correction processing of the amplification
device of the second embodiment. In FIG. 5A, in the listening room, a monitor is installed, the
monitor is in front, two front speakers are installed on the left and right front of the listening
point, and two rear speakers are on the left and right behind the listening point. It is assumed
that it is installed.
[0103]
The measurement points of the frequency characteristic are a listening point P1 when there is
one listener, a point P2 to the left of the listening point P1, a point P3 to the right of the listening
point P1, a point P4 ahead of the listening point P1, It is assumed that a point P5 behind the
listening point P1.
[0104]
For measurement of frequency characteristics at each point, a microphone device as shown in
FIG. 5 (b) is used.
The microphone device picks up the signal output from each speaker by each microphone. By
using the microphone device, it is not necessary to move the microphone to the measurement
position, and measurement of frequency characteristics can be performed, so that the
measurement time can be shortened.
[0105]
The signal processing unit 1 of the amplification device performs measurement and calculation
sequentially from the point P1. First, the frequency characteristic of each speaker at the point P1
08-05-2019
34
is measured by the microphone installed at the point P1, and the correction frequency
characteristic Cp1 is determined by the correction processing described in the first embodiment
based on the measurement result.
[0106]
Next, the frequency characteristic of each speaker at the point P2 is measured by the
microphone installed at the point P2, and the correction frequency characteristic Cp2 is obtained
by the correction processing described in the first embodiment based on the measurement result.
Similarly, the microphones installed at points P3, P4, P5 measure the frequency characteristics of
each speaker at P3, P4, P5 at the points, and the correction frequency characteristics Cp3, Cp4,
Cp5 are determined based on the measurement results. . The obtained frequency characteristics
for correction Cp1 to Cp5 are shown in FIG.
[0107]
After obtaining the correction frequency characteristics Cp1 to Cp5, next, the signal processing
unit 1 of the amplification device obtains an average value for each frequency for the correction
frequency characteristics obtained at the five measurement points shown in FIG. . A parametric
equalizer is formed based on the frequency characteristic for correction Cav obtained as a result
of averaging, and the frequency characteristic is corrected.
[0108]
By this, it is possible to correct to the optimum frequency characteristic in the range surrounded
by the points P2 to P5 around the point P1. Therefore, even if the listener moves the listening
point a little, the listener can listen to the audio signal with the optimal frequency characteristics.
[0109]
In the second embodiment, although the measuring apparatus shown in FIG. 5 (b) is used, after
measuring the frequency characteristic and calculating the frequency characteristic for
correction at each point using one microphone, the following operation is performed. The
listener may move the microphone to the measurement point.
08-05-2019
35
[0110]
An amplifier (AV amplifier) used in a multi-channel audio reproduction system such as a home
theater system must connect a plurality of devices and speakers.
Therefore, many terminals for cable connection of the amplification device are provided. The
listener associates the terminals with the speakers and connects them with cables, but often the
installation location of the speakers is incorrect or the connection between the terminals and the
speakers is incorrect, and the connection error is not noticed at the time of connection. The audio
signal may be listened to by the incorrectly connected speaker. In this case, the optimum
surround effect can not be obtained, and the listener has to rewire the cable and takes time and
effort.
[0111]
In the amplification apparatus of the second embodiment, since the frequency characteristics are
measured at five points around the listening point in the case of one listener, connection and
installation of a plurality of speakers are appropriate using the measurement results. It can be
determined automatically. As a result, connection errors between the speakers and the terminals
and installation errors of the speakers can be eliminated, and an environment where the user can
listen to the audio signal in an appropriate state can be easily constructed.
[0112]
FIG. 7 is a view for explaining the arrangement of measurement points and each speaker in the
second embodiment. In the measurement of the frequency characteristic of the output signal
from each speaker in the second embodiment, the distance to each speaker at each measurement
point is calculated. As shown in FIG. 7, the speakers SP1 to SP8 are arranged.
[0113]
08-05-2019
36
When measured at the point P1, the distance between the point P1 and each of the speakers SP1
to SP8 is possible by impulse response measurement. However, it can not be determined in
which direction the speakers SP1 to SP8 are positioned with respect to the point P1.
[0114]
When measured at point P2 following point P1, the difference between the arrival times of the
measurement signals from the respective speakers determines whether each of the speakers SP1
to SP8 is located on the left or right of the X axis direction with respect to point P1 (listening
point) It can be determined.
[0115]
For example, in the measurement result at the point P1, it is assumed that the distance Ls1p1
between the point P1 and the speaker SP1 and the distance Ls3p1 between the point P1 and the
speaker SP3 are the same at the measurement point P1 (Ls1p1 = Ls3p1).
In the measurement result of the point P2, the distance Ls1p2 between the point P2 and the
speaker SP1 and the distance Ls3p2 between the point P2 and the speaker SP3 are shorter in the
speaker SP1 than in the speaker LP3 (Ls1p2 <Ls3p2). That is, it can be determined that the
speaker SP1 is positioned to the left with respect to the point P1, and the speaker SP3 is
positioned to the right with respect to the point P1.
[0116]
Similarly, by measuring at the point P3, the positional relationship between the speakers
obtained from the measurement results at the points P1 and P2 becomes clearer. For example, it
is assumed that the speaker SP2 is installed directly in front of the point P1. In the previous
measurement, the position of the speaker SP2 is determined only to the right of the speaker SP1
in the X axis direction. However, based on the measurement result at the point P3, it can be
determined that the speaker SP2 is located at least between the speakers SP1 and SP3. In
addition, when the distance between point P2 and point P3 is equal to point P1, the position of
each speaker becomes clear by trigonometry from the measurement results of each of the two
measurement points, and speaker SP2 is Y-axis relative to point P1. It can be determined that it
exists on the front or back of the direction.
08-05-2019
37
[0117]
Furthermore, it becomes clear from the measurement result at the point P4 whether the speakers
SP1 to SP8 are positioned before or after the point P1 in the Y-axis direction. Similarly, from the
measurement result at point P5, the arrangement of each speaker becomes clearer.
[0118]
Based on the distances from the speakers to the microphones obtained at these measurement
points P1 to P5, it can be grasped in which direction front, rear, left, and right the speakers SP1
to SP8 are located with respect to the point P1. Based on the measurement result, when the
speaker is not connected properly or the speaker is not installed in an appropriate place, a
warning may be displayed to the listener using a monitor or a display device built in the device
body. it can. As a result, it is possible to reduce the time or effort required for the listener to
reproduce the audio signal once and to rewire the cable after noticing the wiring error.
[0119]
Also, if it is determined based on the measurement results that the cable wiring is not correct,
even if there is a cable wiring error, the audio signal channel to be automatically output is
switched at the output unit inside the amplification device, so that listening can be made. People
do not need to rewire and can watch movies and music.
[0120]
The figure which shows the structure of one Example of the amplifier of this invention.
FIG. 7 is a diagram for explaining correction processing of automatic sound field correction
processing in the amplification device of the present embodiment. The figure for demonstrating
the calculation regarding the filter factor Q in the amplifier of a present Example. FIG. 6 is a
diagram for explaining an operation regarding a correction value R in the amplification device of
the present embodiment. The figure explaining the measurement point in automatic sound field
amendment processing of the amplification device of a 2nd example of the present invention.
FIG. 13 is a view for explaining processing for obtaining correction frequency characteristics in
08-05-2019
38
automatic sound field correction processing of the amplification device of the second
embodiment. FIG. 7 is a view for explaining the arrangement of measurement points and each
speaker in the second embodiment.
Explanation of sign
[0121]
DESCRIPTION OF SYMBOLS 1 ... Signal processing part, 1a ... Memory, 2 ... Output part, 3 ...
Signal generation part for measurement, 4 ... Sound collection input part, 5 ... Operation part, 6 ...
Control unit 7: Speaker 7a: front left speaker (FL) 7b: front light speaker (FR) 7c: center speaker
(C) 7d: surround left speaker SL), 7e ... surround light speaker (SR), 7f ... subwoofer speaker (SW),
8 ... microphone.
08-05-2019
39
Документ
Категория
Без категории
Просмотров
0
Размер файла
57 Кб
Теги
jp2007043295
1/--страниц
Пожаловаться на содержимое документа